CHAPTER III

METHODOLOGY OF THE STUDY

3.1 INTRODUCTION

This chapter presents the matters related to the methodology of the study. Among the
sub-topics of this chapter are method of the study (3.2) and population and sample of
the study (3.3). The instrument used is focused in section 3.4 and in Section 3.5; the
method of data collection is discussed. Methods of analyzing the data are the focus of
Section 3.6. Section 3.7 presents the research framework and the conclusion ends the
chapter.

3.2 METHOD OF THE STUDY

This is a survey study that focuses on finding out the levels of South Australian
Matriculation’s (SAM) pre-university students for self-directed learning (SDL) and
whether they have different levels of readiness for SDL due to certain individual
differences. This study is also aimed to explore the area of language learning
strategies (LLS) as to find out if there is any relationship between their levels of
readiness in SDL and their LLS levels. This study also attempts to discover the types
of strategies they use in learning English.

The survey method is chosen as it enables the researcher to elicit standard

information regarding the levels of SDL of the selected samples through a series of
standardized questions. This study is based on the quantitative approach, appropriate
with the samples involve in the present research. The study is targeting younger adults
64

of 17 to 18 years old and if a qualitative method is adopted, they may not be able to
justify their answers for the statements in the Self-Directed Learning Readiness Scale
(SDLRS) questionnaire. One reason for this can be their lack of familiarity with SDL
concept.

Survey studies have changed tremendously in the past few years, from a simple
pencil and paper method; they have undergone major technological changes. Still,
survey studies are able to generate valuable information from the respondents either
by using the straightforward combination of pencil and paper or online survey to
collect data (Burns 2000). Survey studies are comparatively inexpensive to conduct
but generally they are able to reach out to a large number of samples and generally
high reliability is easy to obtain. This can be achieved by presenting all subjects with
a standardized stimulus and thus eliminated any observer’s subjectivity.

Another added advantage for survey studies that is suitable for the present
study is the use of standardized questionnaires that are administered to the samples.
Despite the situation of the samples, all of them will receive the same questions, with
consistent or the same definitions for all the questions or terms used in the
questionnaires. The power of standardization in the questions posed to the samples
will ensure that similar data can be collected from all the respondents.

3.3 POPULATION AND SAMPLE OF THE STUDY

The respondents of this study were the South Australian Matriculation (SAM)
students that were taking ESL II course for June – October 2003 semester. They
composed of 186 students from 21 classes of five different majors: Legal Studies,
Business Studies, Information Technology, Biotechnology and Pure Mathematics. For
January 2003 intake, majority of the students tend to enroll for Business Studies and
Biotechnology majors than the other fields and thus the samples tend to be bias in this
matter.

ESL is a compulsory subject for all SAM students. This subject is divided to
ESL I for semester one and ESL II for semester two. A component of ESL II is to
65

expose and familiarize students with International English Language Testing System
(IELTS). Other than the lessons in class, the English Language Centre (ELC) is also
holding IELTS workshops and examination preparatory classes. Most of these
students may opt to further their studies in Australia where IELTS is a requirement.

IELTS is similar to Test of English as a Foreign Language (TOEFL). Most

Australian, British, Canadian and New Zealand academic institutions accept IELTS.
Hence, if they choose to go to one of those countries mentioned, IELTS is also
important to them. TOEFL is a standard English test that measures the ability of non-
native speakers of English to use and understand North American English as it is used
in the college and university settings. Similar to TOEFL, IELTS tests the complete
range of English language skills that will commonly be encountered by students when
studying or training in the medium of English.

The samples were chosen based on random sampling. The researcher had
obtained the assistance of eight other English language teachers to select them. Ten
students were selected from each class based on the name list in the class register. The
students’ registration numbers were written onto slips of paper and then the slips were
shuffle in a container. The teacher drew out one slip at a time and then put it back in
the container and continued to draw the slips until the ten students had been selected.
This simple procedure was done in the presence of the researcher.

Technically, in the initial stage of selecting the samples and collection of data,
210 samples were gathered. Nonetheless, the number reduced as some of the students
left the program either to join other programs or other colleges as some of them were
the recipients of scholarships. In the end, only 186 of the original samples remained.

3.4 INSTRUMENT

In order to assess the students’ levels of readiness for SDL and their choices of LLS,
two questionnaires - Self-Directed Learning Readiness Scale (SDLRS) and Strategy
Inventory for Language Learning (SILL) were used.
66

3.4.1 Self-Directed Learning Readiness Scale (SDLRS)

Self-Directed Learning Readiness Scale (SDLRS) was developed by Lucy M.

Guglielmino through a three-round modified Delphi Survey with a panel of
authorities on self-direction in learning (Barrett 1991). Some of the participants were
Arthur Chickering, Patricia M. Coolican, Cyril O. Houle, Malcolm S. Knowles and
Allen Tough (Barrett 1991). SDLRS has been used to assess the degree to which
individuals themselves perceived to possess certain attitudes, skills and traits
associated with SDL (Long 1987). SDLRS consists of 58 item self-report
questionnaire with Likert-type items. Subjects are expected to indicate how much they
agree with each item on a scale from 1 to 5 (Long 1987). Each number notes certain
measurement as below:

 1 - Always never true of me; I hardly ever feel this way.

 2 – Not often true of me; I feel this way less than half the time.
 3 – Sometimes true of me; I feel this way about half the time.
 4 – Usually true of me; I feel this way more than half the time.
 5 – Almost always true of me; there are very few times when I don’t feel this
way.

Seventeen statements in the questionnaire demonstrate the learners’ negative

behaviour towards SDL and the other 41 statements illustrate the learners’ positive
behaviour towards this type of learning format. The two categories of statements are
all listed and grouped together, thus camouflaging the negative statements among the
positive ones. Nonetheless, the value of positive and negative statements differs.

Statements 3,6,7,9,12,19,20,22,23,29,31,32,35,44,48,53 and 58 are the

negative statements. An example of such statements is “When I see something that I
don’t understand, I stay away from it”. This statement denotes a preference of an
individual who is low in SLD. For each statement in this category that is rated as 1, is
multiplied with 5, statements that are rated as 2, are multiplied with 4, 3 with 3, 4 with
2 and 1 with 5. The score of each statement is then totaled to be labeled as the “R-
Total”.
67

However, for the remaining positive statements, the calculation method is

reversed. The statements rated as 1 are multiplied with 1, statements rated as 2 with 2,
until the statements rated as 5 that will be multiplied with the value of 5. The totals
are then added up to make up the “P Total”. One example of such statements is “If
there is something I want to learn, I can figure out a way to learn it”.

Both the “R Total” and “P Total” are then added to signify the measure of the
students’ levels of readiness in SDL. The students’ scores are interpreted as in the
following table:

TABLE 3.1 Score and Level for SDLRS

SCORE RANGE LEVEL OF READINESS

Below 209 LOW
210 – 238 AVERAGE
239 – above HIGH

Source: Guglielmino 1977 in Norzaini Azman et al.. 2003

The responses to the questionnaire will also indicate whether the respondents
possess these eight factors that outline the prerequisite criteria of self-directed learners
as defined by Guglielmino (1977 in Shokar et al. 2002)

 openness to learning opportunities,

 self-concept as an effective learner,
 initiative and independence in learning,
 informed acceptance of responsibility for one’s own learning,
 love of learning,
 creativity,
 future orientation and ability to use basic study skills and
 problem-solving skills.

High scores indicate persons who prefer to determine their own learning needs
and plan and implement their own learning but it does not mean that they will never
68

choose to be in a structured learning situation. They may well choose traditional

courses or workshops as part of a learning plan (Barrett 1991).

SDLRS has been translated into French, Spanish, Japanese, Chinese, Korean,
German, Finnish, Greek, Italian and Portuguese and in its original form; it has been
used in hundreds of research efforts, including a large number of master’s theses and
doctoral dissertations (Guglielmino and associates 2003). Locally, SDLRS has also
been translated into the Malay language. The fact that SDLRS has been used in
researches worldwide provides a strong support on the validity and reliability of it as
the chosen tool to assess the samples’ levels of readiness for SDL.

SDLRS’s reliability coefficient has been tested in many other studies to ensure
its reliability. The present study’s reliability coefficient was 0.89. Table 3.2 shows the
reliability coefficients of the other studies that used SDLRS.

TABLE 3.2 Reliability Coefficient of SDLRS

Researcher & Year Sample Size SDLRS Reliability

Chang (1991) 1 79 0.91
Tri Darmayati (1994)2 417 0.91
Norzaini (1999) 3 150 0.92
Norzaini (2003)4 334 0.88

Source: 1,3 &4 - Norzaini Azman et al. 2003

2
– Tri Darmayati 1994

The reliability coefficient obtained for this study was slightly lower than the
ones obtained in other studies. Nonetheless, Garrison (1997) has estimated that the
internal reliability for SDLRS has consistently been above 0.85. Based on the
comparison and Garrison’s statement, it is concluded that SDLRS is suitable to be
used with the samples of the present study.
Originally, Guglielmino had administered SDLRS to different groups of
learners of which included ninety-five high school juniors and seniors and a group of
college undergraduates attending daytime classes. Guglielmino characterized the
69

samples as being an acceptable representative of the subjects with which SDLRS was
to be used (Long 1987). The statement above lent confidence that SDLRS was
suitable and appropriate to be used with these young adults and also this has also been
certified in the pilot test done with another group of SAM students.

3.4.2 Strategy Inventory for Language Learning (SILL)

Originally, Rebecca Oxford produced Strategy Inventory for Language Learning

(SILL) as an instrument for assessing the frequency of use of LLS by students at the
Defense Language Institute (Oxford & Burry-Stock 1995). There are several version
of SILL but Version 7.0 of SILL is especially produced for students of other
languages learning English either as a second or foreign language. It has been used
widely to find out the type of LLS that are used by students in learning English
(Oxford 1990).

SILL contains fifty Likert-scale statements related to learning English.

Subjects are expected to rate themselves according to the rating given.

 1 – Never or almost never true of me

 2 – Usually not true of me
 3 – Somewhat true of me
 4 – Usually true of me
 5 – Always or almost always true of me

The fifty statements are sub-divided to six parts and each part represents the
major group of strategies used in learning a language.

 Part A contains nine statements on memory strategies

 Part B contains fourteen statements on cognitive strategies
 Part C contains six statements on compensation strategies
 Part D contains nine statements on metacognitive strategies
 Part E contains six statements on affective strategies
70

 Part F contains six statements on social strategies

The strategies described in the SILL were drawn from a comprehensive

taxonomy of LLS covering the four skills including reading, writing, speaking and
listening (Rosna and Sharifah 1994).

Respondents will get to know the average of each of the strategies and also the
overall average of the six parts of strategies. SILL does not categorize learners as
good, average or poor language learners but measure how frequent learners apply
LLS in learning English.

Like Guglielmino’s SDLRS, SILL has also been translated into several
languages such as Chinese, Korean, Arabic, German, Spanish, Thai, Ukrainian,
Japanese and French (Oxford & Burry-Stock 1995). Oxford (1990) has prepared a
special package for any researcher to use SILL, in which it includes a set of directions
to the students, scoring worksheet, summary profile, strategy graph and background
questionnaire. SILL is also used not to just explore the type or types of students’ LLS
but also acts as an exposure for the students to other ways of learning languages.

There are a number of related questionnaires that have been devised to gauge
students’ use of LLS. They range from a simple one-page questionnaire – What Sort
of Language Learner are You? (Ellis and Sinclair 1989) to a more elaborate one
like Oxford’s SILL. A prominent local researcher in this field – Mohamed Amin
Embi (2000) has also produced a much-sought Strategy Questionnaire that is used to
find out what the students actually do when they learn English.

Muhamed Amin’s questionnaire is notable as it covers three different situations

in which learning of English may take place for second language learners. Based on a
four-point likert scale, respondents are supposed to mark the statements that describe
their way of learning English in class (Classroom Language Learning Strategy –
CLLS), out of class (Out-of-Class Language Learning Strategy – OLLS) and when
preparing for the examination (Exam Language Learning Strategy - ELLS). Other
types of strategy rating scales than SILL are Chamot et al.’ s Learning Strategies
71

Inventory and McGroarty’s Language Learning Strategy Student Questionnaire

(Oxford and Burry-Stock 1995).

SILL Version 7.0 was used in this study as it fulfills the aim of the study of
finding out the types of LLS that were used by the students on the whole. Oxford
(1990) describes six different LLS in SILL and the outcome will provide the answer
of which type of LLS that was most commonly used by the samples. Moreover, it
might also help the samples of this study to try out other LLS that they could use to
improve their English.

SILL’s reliability coefficient has also been tested in many previous studies.
Table 3.3 presents the outcome of some of the studies that utilized SILL as their main
instruments.

TABLE 3.3 Reliability Coefficient of SILL

Researcher & Year Sample Size SILL Reliability

Philips (1990)1 141 0.87
Anderson (1993)2 95 0.91
Talbot (1993)3 31 0.85
Bremmer (1998)4 149 0.92

1,2 & 3
Source: Oxford & Burry-Stock 1995
4
Bremmer 1998

In general, Oxford and Burry-Stock (1995) state that the ESL/EFL SILL’s
reliabilities have been slightly lower than the reliabilities for the translated version of
ESL/EFL SILL version. Nonetheless they added that the reliabilities obtained for
SILL English version are still in the very acceptable range. The reliability coefficient
for this study was 0.87 and it was the same with the reliability coefficient obtained in
Phillips’s studies (1990 in Oxford & Burry-Stock 1995), thus signifying its suitability
to be used as an instrument for this study.

3.4.3 Pilot study

72

Two separate pilot studies were conducted prior to the data collection weeks to
foresee any possible difficulties that might emerge during the real data collection
stage.

The researcher administered both questionnaires to a randomly selected group

of 32 SAM students from two different classes. The respondents were informed that
they had been selected for a pilot study and that it involved two different sets of
questionnaires. The Strategy Inventory for Language Learning (SILL) was given one
week after the administration of Self-Directed Learner Readiness Scale (SDLRS).

The students were briefed on the basic concept of SDL and the aims of the
questionnaire. They were also shown on how to complete the calculation at the end of
the questionnaire. The samples were reminded to seek clarification from the
researcher if there were any questions that were not clear to them. They were
instructed to fill out the questionnaire within the specified time. Overall the students
took fifteen to twenty minutes to complete the questionnaire in addition to calculating
their “P” and “R” total and simultaneously the overall score.

SILL was administered exactly a week after SDLRS’ administration. It began

with a briefing of what SILL is used for and later, a short lecture recalling on the
importance of learning strategies in learning a language. The students were given the
SILL questionnaire with the background questionnaire to collect data on the
particulars of the respondents. In all, the students took about 30 minutes to fill out the
questionnaire and do the calculation to find out their own types of LLS. There were
no questions asked on the statements in the questionnaire.

The SDLRS and SILL used for the pilot study had been tested for their
reliabilities and the results are presented below:
TABLE 3.4: Reliability Result for Pilot Tests

SDLRS SILL
Reliability 0.91 0.87
73

3.5 DATA COLLECTION

The researcher obtained the assistance of eight other English language teachers to
collect the data. The sets of questionnaires were distributed to all selected teachers at
the end of June 2003. The first questionnaire (SDLRS) was given to the students in
the last week of June and SILL was administered in the first week of July.

The researcher contacted each of the English teachers personally in order to

select the time for the questionnaires to be distributed. Generally, the teachers agreed
to distribute the questionnaires during the two-hour period that is normally allocated
for in-class writing session. All the teachers were individually briefed on the
objectives of the study and the nature of the questionnaires. Some of the teachers
involved had also attempted the questionnaires to ensure that they understood the
nature of the questionnaires and to prepare for any possible questions that the students
may ask during the data collection period. Four teachers had also requested the
researcher to drop by the class during the data collection time to assist them in case of
any unforeseen situation.

The selected students were called and briefed on their involvement in the
study. They were informed of the dates the data would be collected and that two
different questionnaires were to be used. As planned, the first questionnaire (SDLRS)
was distributed in the last week of June (23rd - 27th June), which was the second week
after the students had their short semester break. It had been suggested and agreed by
the teachers to give the questionnaires in the last half hour of the class, as it would not
interfere with the lesson.

After the briefing, the questionnaires were distributed to the selected students
and they were instructed on how to answer it. The students were asked to complete
the calculation of marks at the end of both questionnaires as per instruction of the
teachers. The researcher then met the teachers to collect the questionnaires and
discuss of any problems that they had experienced during the data collection period.
74

The second questionnaire (SILL) was planned to be given to the students

exactly a week after the first questionnaire that was the first week of July (2 nd to 6th
July). The same procedure was observed. However, for this time, the students were
requested to complete background questionnaires to collect other important
information for the study. These were the additional information collected:

 gender,
 race,
 school major,
 current academic standing and
 ESL semester one grade.

Customarily, the questionnaires were administered in a group setting for the

convenience of the class teachers and researcher.

3.6 ANALYSIS PROCEDURE

Basically, six types of analyses were used in this study. They were

1. frequency count,
2. mean score,
3. t-test,
4. one way Analysis of Variance (ANOVA),
5. chi-square and
6. cross-tabulation.

For the most part, the data was analyzed using the Statistical Package for
Social Science (SPSS) Version 11 package but the total for both questionnaires were
calculated by the samples. To find out the levels of readiness for SDL, the samples
would calculate the R and the P total separately and later combine the two totals as to
obtain the final score that signified their levels of readiness for SDL. Frequency
75

count was used to find out the number of students that fall in the three categories of
SDLRS.

Each of the averages in the six parts of SILL was also computed by the
samples themselves. To attain the average score of each part of SILL, the respondents
had to total-up the score of each part and then to divide it by the numbers of questions
exist in that particular part. For instance, there are nine questions in Part A, the sample
would add-up her total for Part A and then to divide that particular score with nine
(represents the number of questions in Part A). The obtained score was the average
score for the respondent for Part A (Memory strategies) of SILL. The same method of
calculation applies for all the parts in SILL. The respondents would also have to
calculate the overall score for all the parts in SILL by adding up all the average scores
for Part A to Part F and to divide that score with six (representing the six different
types of LLS). Overall, respondents would obtain seven different scores, representing
Part A to Part F and also the average score for SILL. To answer the research question
as to find out the type of LLS that was most used by the samples, descriptive analysis
was used to find out the mean score for each of the LLS. For this study, recalculations
of the samples’ scores were done to ensure its accuracy.

The computation for all the hypotheses was done by SPSS. To determine the
significance throughout the study, the standard p < 0.05 was used. This means that a
result will be considered statistically significant if it could have occurred by chance
fewer than 5 times out of 100 (Burns 2000). Table 3.4 shows the types of analyses
that were used to test each of the hypotheses.

TABLE 3.5 The Hypothesis, Instrument and Statistical Analysis of the Study

No Research Questions and Hypotheses Instrument Statistical Test

1 RQ: What are the levels of pre-university SDLRS Frequency count
students’ readiness for SDL?
76

2 RQ: Do students of different demographic

background such as gender, race and school
majors have different levels of readiness for
SDL?

Hypothesis 1
Ho: There is no significant difference T-test & cross-
between students’ gender and their levels of tabulation.
readiness for SDL.

Hypothesis 2
Ho: There is no significant difference One way
between students’ race and their levels of ANOVA &
readiness for SDL. cross-tabulation

Hypothesis 3
Ho: There is no significant difference One way
between students’ school majors and their ANOVA &
levels of readiness for SDL. cross-tabulation
3 RQ: What relationship, if any, exists
between students’ current academic standing
and their levels of readiness for SDL?

Hypothesis 4
Ho: There is no significant relationship
between students’ current academic
standings and their levels of readiness for
SDL.
Chi-square &
Hypothesis 5 cross-tabulation
Ho: There is no significant relationship
between students’ ESL results and their
levels of readiness for SDL.
Chi-square &
cross-tabulation
4 RQ: Which type of LLS, of the six SILL Frequency and
categories, that is mostly used by the pre- mean score
university students and what is their overall
mean?
5 RQ: What are the students’ current levels of Mean score
LLS?
6 RQ: Do students of different gender and race
use different types of LLS?

Hypothesis 6
Ho: There is no significant difference
T-Test
between students’ gender and the types of
LLS used.
77

Hypothesis 7
Ho: There is no significant difference
between students’ race and the types of LLS One way
used. ANOVA
7 RQ: What relationship, if any, exists
between students’ ESL results and their SILL
scores?

Hypothesis 8
Ho: There is no significant relationship Chi-square &
between students’ ESL result and their SILL cross-tabulation
scores.
8 RQ: What relationship, if any, exists
between students’ SILL and SDLRS scores?

Hypothesis 9
Ho: There is no significant relationship SDLRS and Chi-square &
between students’ SILL and SDLRS scores. SILL cross-tabulation

3.7 RESEARCH FRAMEWORK

The following model illustrates the research framework that was used as the basis of
conduct of this study.

Two types of questionnaires are used in this study and each measures different
characteristics in learners. SDLRS is used to measure the levels of readiness of the
respondents in SDL and SILL is utilized to gauge the frequency of learners using
LLS. There are two independent variables that are tested to see the significance of
them against SDLRS and SILL: gender and race.

There were other research questions and hypotheses formulated to test several
independent variables against SDLRS and SILL. For the former, these elements are
researched: levels of readiness for SDL, its relationship with school majors and
current academic standings and its relationship with samples’ ESL scores. SILL is
employed to study or find out the following elements and variables: type/s of LLS
used, their overall mean, current levels of LLS and the relationship between SILL and
ESL scores. The final hypothesis is to test the relationship between the samples’
levels in SILL and SDLRS.
78

The aim of both learning formats is to produce active learners that are
responsible and capable to conduct their own learning with or without the assistance
of others. Generally, self-directed learners are described as being able to direct
themselves in learning and thus, with that ability they should be able to direct
themselves in learning English.

SDLRS SILL
shared variables
gender
race
Other tested Other
variables HIGH HIGH tested
 levels of variables
readiness AVERAGE MEDIUM  types of
 school LLS
majors LOW LOW used
 current  overall
academic mean
standing  current
 ESL vs. user
SDLRS levels
 ESL vs.
relationship

HSDL = HLLS

 HSDL – high category self-directed learning

 HLLS – high user of language learning strategies

FIGURE 3.1 Research Framework

3.8 CONCLUSION

This chapter centers on the method used for this study and the samples that were
involved in seeking the answers to the research questions. The two different sets of
questionnaires used as the instruments are discussed in Section 3.4. The analyses that
are used to analyze the data collected are the focus of Section 3.6. Basically, the SPSS
79

package is used to analyze each of the data but there are some basic calculations done
by the samples themselves. Each of the calculations done by the samples has been
recalculated to ensure their accuracy and validity.