THE CORRELATION BETWEEN SOCIO-ECONOMIC STATUS AND LANGUAGE LEARNING STRATEGIES OF EFL/ESL LEARNERS.

THE CORRELATION BETWEEN

SOCIO-ECONOMIC STATUS AND LANGUAGE

LEARNING STRATEGIES OF EFL/ESL

LEARNERS

THESIS

Submitted in partial fulfillment of requirement for the degree of

Sarjana Pendidikan (S. Pd) in Teaching English

By:

Khavit Syaifudin Sharp

NIM D05212015

ENGLISH EDUCATION DEPARTMENT

FACULTY OF TARBIYAH AND TEACHERS TRAINING

SUNAN AMPEL STATE ISLAMIC UNIVERSITY

SURABAYA

2016

THE CORRELATION BETWEEN

SOCIO-ECONOMIC STATUS AND LANGUAGE

LEARNING STRATEGIES OF EFL/ESL

LEARNERS

THESIS

Submitted in partial fulfillment of requirement for the degree of

Sarjana Pendidikan (S. Pd) in Teaching English

By:

Khavit Syaifudin Sharp

NIM D05212015

ENGLISH EDUCATION DEPARTMENT

FACULTY OF TARBIYAH AND TEACHERS TRAINING

SUNAN AMPEL STATE ISLAMIC UNIVERSITY

SURABAYA

2016

ii

APPROVAL SHEET

This thesis by Khavit Syaifudin Sharp entitled “The Correlation Between Socio-Economic Status and Language Learning Strategy of EFL/ESL Learners” has been approved by the thesis advisor for further approval by the Boards of the Examiners.

Surabaya, August, 12th ₂₀₁₆

This thesis by Khavit Syaifudin Sharp entitled “The Correlation Between Socio-Economic Status and Language Learning Strategy of EFL/ESL Learners” has been examined on August, 15th _{2016 and approved by the board of examiners.}

Dean,

(Prof. Dr. H. Ali Mudlofir, M.Ag) NIP. 197408042000031002

Chair,

(M. Hanafi, MA) NIP. 197408042000031002

Secretary,

(Rizka Safriyani, M.Pd) NIP. 198409142009122005

Examiner I,

(M.Syaifudin, M.Ed, Ph.D) NIP. 197310131997031002

Examiner II

(Dr.M.Salik, M.Ag) NIP. 196712121994031002

iv

APPROVAL SHEET

With the grace and love of Allah, the thesis by Khavit Syaifudin Sharp entitled “The Correlation Between Socio-Economic Status and Language Learning Strategies of EFL/ESL Learners” was dedicated for:

My beloved parents; Abdul Hamid and Siti Khusnul Khotimah who always give their love, kindness, attention, pray, and the value of life, so I have a big responsibility for all of duty and all of obligation that I have to complete.

My best friends, Ermaris Suweni, Fitra Amalia, Hafiz Zidnin Alfarizi, Hanifatul Ummah, Lailatul Mardiyah, Nurul Huril Aini, Siska Aprilianti, Siti Maghfirotun Hasaniyah, and Wahyu Hidayatulloh, who always give a spirit, a story, and a new experience. So I always have an idea to solve the problem when I was writing this thesis.

Nur Aifa Rosyidah, someone with a high spirit and motivation, she always could make me smile with her funny talking style, so I could break from my tiredness in thesis writing process. And one important thing I have been learned from her was “don’t give up, just smile and do it, this life will not fun without a problem as our friend.”

The special one who always in the deep in my heart, someone that I really love, Azzarin Valda Adilla. She always cares and smile to me, that is make a source of spirit and unlimited braveness inside my heart, so I could take a right decision in the right time. Thank for the experience, because of that, I could be a better person, have more responsibility, open minded, smarter to take a decision, and more appreciate to the care of other person.

ABSTRACT

Sharp S. Syaifudin (2016). The Correlation Between Socio-Economic Status and Language Learning Strategies of EFL/ESL Learners. A Thesis. English Education Department, Faculty of Tarbiyah and Teacher Training, Sunan Ampel State Islamic University, Surabaya. Advisors: M. Hanafi, M.Ag, MA

Key Words: Socio-Economic Status, Language Learning Strategies (LLS)

Language learning strategies (LLS) and Socio-Economic status (SES) was factor form many that had correlation and affected the academic achievement or academic performance. It was many studies asserted that LLS had a strong positive correlation and affected the result of academic purposed. According to some research, SES also indicated had a positive correlation with academic purposed. The focus of this study was to investigate the correlation between Socio-Economic status and language learning strategies. The correlational method was used because this research was the first study which was investigate SES and LLS. This study was a guidance for the further research to improve the quality of learning strategy by overcame the differences level between Socio-Economic status. 75 learners of English Education Department of Islamic University of Sunan Ampel Surabaya were participated. The respondents were chosen using simple random sampling method. The MANOVA and Pearson Correlation was carried out to investigate the relationship between Socio-Economic status and language learning strategies. The result of the study asserted that there was a strongly positive linier correlation between Socio-Economic status and language learning strategy.

viii

TABLE OF CONTENT

APPROVAL SHEET ... ii

APRROVAL SHEET ... ii

DEDICATION SHEET ... iii

MOTTO ... iv

ACKNOWLEDGEMENT ... v

ABSTRACT ... vi

PERNYATAAN KEASLIAN TULISAN ... vii

TABLE OF CONTENT ... viii

LIST OF TABLE ... xi

LIST OF FIGURES/GRAPICS ... xii

LIST OF APPENDICES ... xiii

CHAPTER I: INTRODUCTION A. Research Background... 1

B. Research Question ... 4

C. Objective of the Study ... 4

D. Hypothesis ... 4

E. Significant of the Study ... 4

F. Scope and Limitation ... 5

G. Definition of Key Term ... 6

1. Socio-Economic Status (SES) ... 6

CHAPTER II: REVIEW OF RELATED LITERATURE

A. Review of Related Literature ... 10

1. Socio-Economic Status (SES) ... 10

2. Language Learning Strategies (LLS) ... 15

B. Review of Previous Study ... 19

CHAPTER III: RESEARCH METHOD A. Research Design ... 24

B. Population and Sample ... 26

1. Population ... 26

2. Sample ... 27

C. Research Instrument ... 28

1. Language Learning Strategies Instrument ... 28

2. Socio-Economic Status Instrument ... 29

a. Pilot Validity 1 ... 30

b. Pilot Validity 2 ... 32

c. Pilot Validity 3 ... 33

d. The Reliability Test ... 35

D. Data Collection Technique ... 35

E. Data Analysis Technique ... 36

1. Descriptive Statistic ... 38

a. Mean ... 39

b. Median ... 40

c. Modus ... 40

d. Standard Deviation (SD) ... 40

e. Z Score ... 40

f. SES Classification ... 41

2. Inferential Statistic ... 41

a. Normality Test/ Normal Distribution test (Kolmogorov Smirnov Test) ... 42

b. Homogeny Test (Levene’s Test) ... 43

c. MANOVA ... 44

d. Pearson Correlation Product Moment ... 45

CHAPTER IV: FINDING AND DISCUSSION A. Finding ... 47

1. The Learners Classification According to their SES Level ... 47

2. Normality Assumption ... 48

3. Homogeny Test ... 49

x

a. MANOVA ... 51

b. Between Subject-Effect Test ... 51

5. Pearson Correlation Product Moment ... 52

B. Discussion ... 55

CHAPTER V: CONCLUSION AND SUGGESTION A. Conclusion ... 61

B. Suggestion ... 61

REFERENCES ... 63

CHAPTER 1

INTRODUCTION

A. Research Background

An assumption in the learning process would give the big effect on learning achievement. Learners who thinking they could attain a good achievement in learning, they would try with their own way to get it. On the contrary, learners who thinking that good learning achievement only could attain by learners with a specific factor, they had no motivation to achieve it.

According to the researcher experience when to carry out PPL at MA Nurul Huda and EYL observation in MI Wachid Hasjim, some students with the mindset that English only could be learned and mastered by students who had the good Socio-Economic background were found. Moreover, at the college level, most of the learners had equal mindsets. For example, they who came from high Socio-Economic family or well-educated parents had a good achievement in English because they had an additional course. This phenomenon also occurred in college learners and parents, it had been known after the researcher asked an opinion to some people about that case.

Some of the research asserted that Socio-Economic background had effects on learning achievement. Parents education was an important factor in Socio-Economic that affect the learners’ performance and learners’ achievement. It was mean, there was a positive correlation

2

between parents’ education with the students developed in learning.1 _Other research presented that learners with high Socio-Economic status (SES) had better education achievement then learners with low SES.2

However High SES give a significant contribution to the quality of learning achievement or performance, the fact points out that Socio-Economic background was not everything to get good learning achievement, it was more depended on the learners.

Some researcher argued that the best learning method depended on the learner itself. Nyikos and Oxford stated that learning begins with the learners.3 _{It was supported by the result of the study which presented that} there was no method guaranteed the successful of learning. The learner may have the best teacher with the amazing method, but when they could not accept it, the method would be useless. Because of this reason, researchers shifted their focus into the learners itself. They think that the learners were more know about the learning method which was acceptable by their selves. Researchers named it as learning strategy, while in language learning, it was known as language learning strategy (LLS).

Some studies presented that LLS had a significant contribution to language learning achievement. The successful language learners had

1 _{Ralph C. Wilson, “Ls There a Positive Correlation between Socioeconomic Status and Academic} Achievement?” (St. John Fisher College, 2008), 12–20.

2 _{Daniel H. Caro, “Family Socioeconomic Status and Inequality of Opportunity” (Freie Universität} Berlin, 2010), 122.

3 _{Carol Griffiths and A. I. S. St Helens, Language Learning Strategies: Theory and Research (AIS} St Helens, Centre for Research in International Education, 2004), 2.

3

more learning strategies used then less-successful language learners.4 Another study also asserted that language learning strategy had an effect on language proficiency.5

According to that reason, it could be concluded that SES and LLS had effects on learning achievement. It became the interesting reason in this research performance. Language learning strategy was a method developed by each learner to make an effective and enjoyable learning. The researcher assumed that the LLS was affected by another factor. Because the learning method came from adopting the other learner strategies then it developed into the new learning method owned by each learner. SES probably was indicated as one factor that affecting LLS, because according to some researchers, both of them had an equal effect on learning achievement.

This study investigated the correlation between SES and LLS with 75 participants of English Education Department on Faculty of Tarbiyah and Teacher Training in Sunan Ampel State Islamic University as the subject. A multivariate analysis (MANOVA) was used to reveal there was any correlation between SES and LLS or not. While Pearson correlation product moment was used to check out the kind of the correlation.

4 _{Pezhman Zare, “Language Learning Strategies among EFL/ESL Learners: A Review of} Literature,” International Journal of Humanities and Social Science 2, no. 5 (2012): 167.

5 _{Norman Fewell, “Language Learning Strategies and English Language Proficiency: An} Investigation of Japanese EFL University Students,” TESOL Journal 2 (June 2010): 159.

4

B. Research Question

What is the correlation between the SES (Socio-Economic Status) and LLS (language learning strategies) of EFL or ESL learners in Sunan Ampel State Islamic University?

C. Objective of the Study

To reveal the correlation between the SES (Socio-Economic Status) and LLS (language learning strategies) of EFL or ESL learners in Sunan Ampel State Islamic University.

D. Hypothesis

��= There was no relationship between Socio-Economic status and language learning strategies

�1= There was a relationship between Socio-Economic status and language learning strategies

E. Significant of study

The research had been done was purposed to give benefits to some people especially which correlating with the educational aspects. Specifically, it was explained bellow:

1. For Teacher

With the result of this study, the teacher could design appropriate learning method according to the Socio-Economic background and language learning strategy of the learners.

5

2. For Students

They can design their own language learning strategies according to their own Socio-Economic background, so it can be more effective and impressively unimposing their condition.

3. For the Researcher

For the further research, the researcher can develop this study deeply. Because this study only investigates the correlation of Socio-Economic status and language learning strategy generally. While for the further research, the researcher could investigate the effect of SES in every aspect of learning strategies. In addition, the researcher in future could investigate to create an effective learning strategy by removing or undercover the SES level differences.

F. Scope and Limitation of the Study

This study was aimed to determine the correlation between SES (Socio-Economic Status) and LLS (Language Learning Strategy). Learners’ family Income, occupation, and education were an indicator to measure the SES value.

While for language learning strategy, there were six strategies; memory strategies, cognitive strategies, metacognitive strategies, social strategies, compensation strategies, and affective strategies as factors used as a measurement tool. Those mentioned factors did not investigate in detail. it was hoped that in the further research the detail of them would be investigated more.

6

The population of this research was all EFL and ESL learners in English Education Department of Faculty of Tarbiyah and Teacher Training in Sunan Ampel State Islamic University. in the sampling technique, the Simple random sampling was used. According to that sampling technique, the sample was taken randomly from the population.6 75 respondents was taken from the population.

The variables of this study were SES (independent) and LLS (dependent), they were natural variable because the researcher did not modify both of them. According to this reason, this research was concluded as survey study. The descriptive quantitative approach was used in data analyzation.

G. Definition of Key Terms

1. Socio-Economic Status (SES)

The study of Socio-Economic status correlated with education was limited, but there were definitions asserted in some of the studies. Socio-Economic status was a term referring to prestige according to the measure of Socio-Economic position as determined by ranking in the social hierarchy.7 _{It was defined as the combination of social status}

6 _{R. Gunawan Sudarmanto, Statistik Terapan Berbasis Komputer: Dengan Program IBM SPSS}

Statistik 19 (Jakarti: Mitra Wacana Media, 2013), 40.

7 _{Brent M. Blevins, Effects of Socioeconomic Status on Academic Performance in Missouri Public}

Schools (Lindenwood University, 2009), 14, accessed June 17, 2016,

7

and economic level which created a position admitted by society. In detail, it was gotten by people with their experience and hard work.8

Socio-Economic status was categorized into three classes; low, medium, and high. The measuring tools were needed to classify SES in that categorization. But because the dimension of Socio-Economic status was large, there were many different features used to measure it. in addition, the tool for identifying the characteristic of Socio-Economic status had to be appropriated with the dimension that would learn.

Several criteria were used to measure Socio-Economic status such as parental income and their profession.9 _{Another research} asserted there was some indication of SES they are income, occupation, and education.10 _{PISA designed indexes to measure} Socio-Economic status based on three components; home possession, educational attainment of parents, and occupational status of parents.11 From those research, most of the criterion of SES was designed based on parental occupation, education, and income. In the another study was also indicated that parents’ occupation, income, and education had

8 _{Charles Kombo Okioga, “The Impact of Students’ Socio-Economic Background on Academic} Performance in Universities, a Case of Students in Kisii University College,” American International Journal of Social Science 2, no. 2 (2013): 38.

9 _{Renato G.Carvalho and Rosa F.Novo, “Family Socioeconomic Status and Student Adaptation to} School Life: Looking Beyond Grades,” Kenneth N.Ross 10, no. 28 (2012): 1212.

10 _{Okioga, “The Impact of Students’ Socio-Economic Background on Academic Performance in} Universities, a Case of Students in Kisii University College,” 38.

11 _{Charles D. Cowan et al., “Improving the Measurement of Socioeconomic Status for the National} Assessment of Educational Progress: A Theoritical Foundation,” National Center for Education Statistics, November 2012, 12.

8

been used as measurement indicator which correlated with educational aspects (learning achievement).12

Family or parents factor always be mentioned in some research correlated with Socio-Economic status. it was mean that that aspect was often used as measurement tool. In most educational research family’ SES background included income, occupation, education, and every yield come from those main aspects frequently used and mentioned as the measurement scale of SES.13

From statements above, it could be concluded that Income, Occupation, and Education especially of the parents was and aspects used to measure the students’ Socio-Economic background.

2. Language Learning Strategies (LLS)

Learning strategies was an effective way of learning which tailored to the capabilities of each individual based on the character, needs, and willingness they had. Chamot asserted that Learning strategies were the conscious thoughts and actions that learners took in order to achieve a learning goal.14 _{But the most basic} definition of language learning was the definition put forward by Oxford, it mentioned that the learning strategy was a step taken by learners to enhance their learning.15 _{The term of language learning strategies came from the combination of} “language” and “learning strategies”. So the language learning strategies (LLS)

12 _{Okioga, “The Impact of Students’ Socio-Economic Background on Academic Performance in} Universities, a Case of Students in Kisii University College,” 38.

13 _{Åse Hansson and Jan-Eric Gustafsson, Measurement Invariance of Socioeconomic Status across}

Migrational Background (University of Gothenburg, June 12, 2012), 1.

14 _{Anna Uhl Chamot, “Issues in Language Learning Strategy Research and Teaching,” Electronic}

journal of foreign language teaching 1, no. 1 (2004): 14.

15 _{Rebecca L. Oxford, Language Learning Strategy: What Every Teacher Should Know, 1st ed.} (Boston: Heinle & Heinle Publisher, 1990), 1.

9

was the effective learning applied by learners in the language learning. LLS contained some features used to identify the LLS which was applied by learners. These features were the factors stimulate learners’ LLS in communicative competence. Metacognitive strategy, affective strategy, social strategy, cognitive strategy, memory strategy, and compensation strategy were the taxonomy in language learning strategy.16 _{Those following feature could be broken down into} more specific aspects, but in this study, the dimension had been investigated was only limited on that general features.

10 CHAPTER II

REVIEW OF RELATED LITERATURE

The review of the related literature was intended to give an evidence and about the conceptual framework and description of Socio-Economic status and language learning strategies. This theoretical explanation was the foundation of this research. On this chapter there were some main aspects were studied, they are the explanation of Socio-Economic status and Language learning strategies.

A. Review of Related Literature

1. Socio-Economic Status (SES)

Socio-Economic status was a measurement of people perception that had correlation with economic and social life. In simply, Socio-Economic status was derived from 2 variables, they are social status and economic status. The position in society was an important factor to acquire social status, while money or the economic quality of the person was associated with economic status.

Marmot & Michael presented (as cited by Okioga) that Socio-Economic status was a combination of economic and sociological aspect.1 _{It was derived from an individual's or family’s economic, a} person's work experience, and social position. In addition, Stephanie and Deborah (as cited by Akhtar) asserted that Socio-Economic status

1 _{Okioga, “The Impact of Students’ Socio-Economic Background on Academic Performance in} Universities, a Case of Students in Kisii University College,” 38.

11

naturally was argued to distinguish between people’s relative position.2 Family income, political power, educational background and occupational prestige3 _{were the dimensions used as the formula of that} differentiation.

According to the statement above, the natural function of SES was classifying the people position in society according to the combination of economic and social status. Generally Socio-Economic status was divided into 3 classes; low, medium, and high. This classification came from the mind-set that was developed in the society. They assumed that the level of education, the total of money or wealth, and the prestigious position was the absolute references of that classification determination.

Annette (as cited by Akhtar) described that income, education, and occupation was the core of SES categorization.4 _{Dickinson &} Adelson also had the same argument that a combination of income, occupation, and education was the essence of SES measurement.5 Their function was to assess the relative status of individual or family.6 In some articles and journals, health and the environment of people included as the tool or aspect to asses Socio-Economic status beside

2 _{Zarina Akhtar, “Socio-Economic Status Factors Effecting the Students Achievement: A} Predictive Study,” International Journal of Social Sciences And Education 2, no. 1 (2012): 281.

3 _Ibid. 4 _Ibid.

5 _{Emily R. Dickinson and Jill L. Adelson, “Exploring the Limitations of Measures of Students’} Socioeconomic Status (SES),” Practical Assessment, Research & Evaluation 19, no. 1 (2014): 2.

6 _{Okioga, “The Impact of Students’ Socio-Economic Background on Academic Performance in} Universities, a Case of Students in Kisii University College,” 38.

12

that three main aspect. But the field of those article is not in education but in political, health, or the other areas.

There are many tool or aspect used to measure SES, it was developed according to the area of the study or research. But in the educational field especially for learners, Socio-Economic background of family or parents became the main concept to measured, because the learners could not create their own status in the society. They got their status according to the family economic and social condition. When the family have good income and good position, people would label the learners as the person who had good Socio-Economic status. Even thought, the family’s or parents’ Socio-Economic background still measured using the three basic aspect (income, occupation, and education)

Beside the three basic concept, the Australian government in their forum paper asserted that Socio-Economic status could be measured on four dimensions, they are educational attainment, occupation, economic resources, and culture or Community.7 _{It can be} applied for all of person (not only for learner)

While Aggarwal et al explained in more detail about the measurement of socio-economic status of the student. By holding on the principle that learners got their Socio-Economic status from their family or parents, they broke down the main indicators of

7 _{Australian Government, “Measuring The Socio-Economic Status of Higher Education Students”} (Department of Education, Employment and Workplace Relation, 2009), 3.

13

Economic status measurement into some criterions based on the family aspects. They are occupation of family head, education of family head, the level of social participation of family head, landholding, housing, farm power, material possessions and type of family which are taken under the prevailing culture.8 _{However those criterions were only the} development of income, occupation, and education in the family side.

In sum, socio-economic status could be measured through three items; education, occupation, and income. While, in educational field especially for learners, those three items were calculated based on the family’s or parents’ condition. This three-dimensional was chosen because it had a correlation each other and could not be separated. Grusky and Weeden asserted that relative position gained by people according to their experience which correlated with education, income, and occupation was always create their natural correlation, one aspect to another was a cause effect relationship.9

8 _{O. P. Aggarwal et al., “A New Instrument (Scale) for Measuring the Socioeconomic Status of a} Family: Preliminary Study,” Indian Journal of Community Medicine 30, no. 4 (2005): 111.

9 _{Dickinson and Adelson, “Exploring the Limitations of Measures of Students’ Socioeconomic} Status (SES),” 2.

Income

Education Occupation

14

For example, income influenced the quality of education a person, then the quality of education influenced occupation, and ultimately, occupation affects income. someone get all that then they would be rewarded or was deemed to have an established social status in society.

Klein (as cited by Dickinson & Adelson) asserted that Income was one of the factors that affect the quality of education and social position in society.10 _{Meanwhile, the occupation also had close links} with income. People who had a good occupation, surely had quite encouraging income. It was mean good occupation also support income, especially for parents and education for students. Horvath, Weininger, and Lareau (as cited by Dickinson & Adelson) stated that the good occupation owned by a person would help him or her to have a connection with the people who are in the prestigious position, such as headmaster of college, the best teacher, director, etc. With that link, the information related to high quality of educational would be obtained.11

Kean (as cited by Dickinson & Adelson) stated that a person with a higher education would be able to make the interaction with people to level up their Socio-Economic status.12 _{They could make a} connection to the business community to support the occupation and

10 _{Ibid., 3.} 11 _Ibid. 12 _Ibid.

15

their income, and could communicate with the teacher or the owner of the foundation to support their education to the higher stages.

Although many factors that could be used as a dimension of the SES, the three factors (occupation, income, and education) were the main measuring instrument and were most often used to measure SES. From those main indicator, the measurement features of SES still could be developed according to the aspect that would be learned. But in educational aspect, Family background with main indicators included would be used to investigates the SES of learners. Parental occupation, total income of the family, the position in the village or an organization and the others were some item examples would be guiding to know about the Family’ SES background of learners.13

2. Language Learning Strategies (LLS)

In the terminology, “learning” was defined as the effective way to learn language independently and fast, it would be developed according to the learners’ behaviour. Furthermore, In the education, especially in language aspect, the “strategy” was interpreted as the effective way to reach the learning achievement. the strategy in learning fulfils these conditions; make learning easier, faster, more self-directed, more effective, more enjoyable, and more transferable to new situations.14

13 _{O. P. Aggarwal et al., “A New Instrument (scale) for Measuring the Socioeconomic Status of a} Family: Preliminary Study,” Indian Journal of Community Medicine 30, no. 4 (2005): 11–12.

16

generally, Learning strategy was defined as the most effective way discovered by the learners to learn according to their habitual, or it could be explained as the appropriate or effective method was appeared on learners to learn according to their need, habitual, want, and demand. O’Malley and Chamot definite SES as “the special thoughts or behaviours that individuals use to help them comprehend, learn or retain new information".15 Oxford asserted that learning strategies was the steps or manners were applied by learners for upgrading their own learning.16

Learning strategy was an essence of learning. Rebecca presented that strategy was important for language learning because it could develop the learners’ communicative competence independently.17 _{With language learning strategy learners enable to} become more independent, autonomous, lifelong learners. In sum, the good language learning strategy can help the learners to learn effectively and independently.

The researcher developed a few characteristics to measure language learning strategy. The characteristic called the Language Learning Taxonomies. some scientists classify LLS into several

15 _{J. Michael O’Malley and Anna Uhl Chamot, Learning Strategies in Second Language}

Acquisition, The Cambridge applied linguistics series (Cambridge [England]; New York:

Cambridge University Press, 1990), 1.

16 _{Rebecca L. Oxford, Language Learning Strategy, 1.} 17 _Ibid.

17

categories. Rubin's (as cited by Gürsoy) argues that LLS has two primary strategic aspects.18

First was the direct strategy consist of clarification / verification, monitoring, memorization, guessing / inductive inferencing, deductive reasoning, and practice.19 _{The second was} indirect strategy, it was consisted of creating opportunities for practice and production tricks.20

Bialystok (as cited by Gürsoy) classifies language learning

taxonomy based on the model of Second language learning found by her. Her classification includes four categories; formal practicing, practicing functional, inferencing, and monitoring.21

But, the most often used to measure language learning strategy was taxonomy designed by Oxford. Through some of the reviews from previous taxonomy, she developed new category of language learning strategy, and it was widely used till today.

Oxford categorized language learning strategies’ core into two categories; direct and indirect strategies. It was equal with Rubins’ taxonomy but sub content in it was different. Direct strategies involved the direct use of language, so it could affect the language learning

18 _{Esim Gürsoy, “Investigating Language Learning Strategies of EFL Children for the} Development of a Taxonomy,” English Language Teaching 3, no. 3 (2010): 166.

19 _Ibid. 20 _Ibid. 21 _Ibid.

18

directly.22 _{whereas she defined indirect strategies as the support of} language learning.23

She classifies the direct strategy into three parts; memory, cognitive, and compensation strategies, while metacognitive, affective and social strategies under indirect strategies.24 _{for more simple please} see figure 2.1.

Figure 2.1 (LLS Taxonomy)

Oxford asserted that direct strategies were like the worker who perform the job.25 _{It was linked to its function, they are to produce the} language which was consist of some strategies. The function of memory strategies is retrieving and remembering new information. While for cognitive strategies is to understanding and producing language. and the function of compensation strategies is for using despite knowledge gaps.

22 _{Rebecca L. Oxford, Language Learning Strategy, 15–16.} 23 _Ibid.

24 _Ibid. 25 _{Ibid., 15.}

19

In addition, Oxford also asserted that indirect strategies were like a director in a job. Its function was to manage how the language was produced.26 _{Equal with direct strategies, it also had sub strategies.} Metacognitive strategies used for coordinating the learning process, affective strategies are used for regulating emotion, and social strategies is used for learning with others.

Direct and indirect strategies must be combined to produce the best possible outcome of language. it was like a workers or performers who need the director to guide them to produce a language.

Those features of taxonomy designed by Oxford would be a tools to measure the language learning strategy in the present study because it has been widely used in several research.

B. Review of Previous Study

In 1995, the study about the Socio-Economic status which was related to education was conferenced by Hirth and Mitchell. The study investigated the relationship between learning achievement which was measured by students score and educational policy making which was correlated with Economic status. It was found that the Socio-Economic status had high correlation with learning achievement of Indiana learners.27 _{The study was explained that Socio-Economic status}

26 _{Ibid., 16.}

27 _{Marilyn A. Hirth and Richard C. Mitchell, “An Indiana Investigation of the Impact of} Expenditures & Socioeconomic Status on Student Achievement: Does Money Matter?” (presented at the National Council of Professors of FAucational Administration Annual Summer Conference, Williamsburg: THE EDUCATIONAL RESOURCES INFORMATION CENTER (ERIC), 1995), 3.

20

was an important factor for learners to get the source access of information to gain the effective and optimal achievement. But because the differences of SES between learners, the source could not be accessed equally by all learners.28

In 2003, another conference namely "Longitudinal Survey of Australian Youth (LSAY)” which was held by Rothman. The focus of the study was to investigate effect of Socio-Economic status and the change of it periodically. the indigenous and non-indigenous learners in Australia was the focus subject of this study. The result was indicated that Socio-Economic influenced the developed the learning achievement significantly. The learners who lived in high Socio-Economic zone had a better access on the source of knowledge. Clearly, the learners who lived in non-metropolitan zone had a lover average of learning achievement than they were lived in metropolitan zone.29 _{The indigenous learners also had} score of learning achievement above the average.

It could be sum that, the location of the learners which affected the Socio-Economic status of the family give an impact on learning achievement. Most of non-indigenous lived in metropolitan location, they got an enough access for their study. While for the indigenous which majority lived in non-metropolitan location, they had less of access on education source.

28 _{Ibid., 15.}

29 _{Sheldon Rothman, “The Changing Influence of Socioeconomic Status on Student Achievement:} Recent Evidence from Australia,” in The Changing Influence of Socioeconomic Status (presented

at the Longitudinal Surveys of Australian Youth (LSAY), Melbourne, Australia: Australian Council for Educational Research (ACEReSearch), 2004), 10.

21

The study was held by Tiffany also supported that there was a positive linear correlation between Socio-Economic status and academic or learning achievement. She also gave the further explanation that family with low Socio-Economic status has less time and attention to their children,30 _{they more focus on working and how the way their family} could eat. While for the high Socio-Economic family, they aware that their children education was more important, so they give more time, attention, and give them a way and guidance to get a better education.

In another study was held by Caro, the Socio-Economic had a significant effect on the learning achievement trajectories between low, mid, and high SES,31 _{it was caused that children more often spend their} time on their environment (informal environment, such as friends and especially their family).

In Pakistan culture, the 13-15 teenagers were abandoned to drive a vehicle.32 _{It was the rule of the parents to give education in order to their} children could have a good learning achievement. She found that had a car or vehicle for children or teenagers gave a negative effect on their achievement.33

30 _{Tiffany Quagliata, “Ls There a Positive Correlation between Socioeconomic Status and} Academic Achievement?” (St. John Fisher College, 2008), 20.

31 _{Daniel H. Caro, “Socio‐economic Status and Academic Achievement Trajectories from} Childhood to Adolescence,” CANADIAN JOURNAL OF EDUCATION 32, no. 3 (2009): 580.

32 _{Zarina Akhtar, “Socio-Economic Status Factors Effecting the Students Achievement: A} Predictive Study,” International Journal of Social Sciences and Education 2, no. 1 (January 2012):

286. 33 _Ibid.

22

That statement was supported by the result of the study which was carried out by Ahmar & Anwar. The subject of this study was six school of eastern united states. They found that majority learners from high SES often got rewarded for their academic achievement, the opposite majority learners especially male, often got punishment.

From the definition and fact above, that could be conclude that Socio-Economic status has positive correlation and affect the academic achievement of the learners based on the process passed by the learners. The family and parents was the most important factors affecting it. With high Socio-Economic status background of family, the learners had more easy way to get an access for their education. Barry stated that Socio-Economic status was an important effect that supporting learners to get access for their study.34

Sirin asserted that family’ Socio-Economic background was the strongest factor affected the learning progress. In many research, family was used as the measurement tool of SES.35 _{Usually it was measured by} income, occupation and education.36 _{However, in small number of} research, a characteristic of neighbourhood and the items had by family also used as indicator.37

34 _{Jennifer Barry, “The Effect of Socio-Economic Status On Academic Achievement” (Wichita} State University, 2006), 25.

35 _{Selcuk R. Sirin, “Socioeconomic Status and Academic Achievement: A Meta-Analytic Review} of Research,” Review of Educational Research 75, no. 3 (Fall 2005): 438.

36 _{Ibid., 439.} 37 _{Ibid., 349.}

23

In the other branch of research, language learning strategy was indicated as the most important factor which was affecting the academic performance or achievement.

The study was conducted by Suwarnak indicate that the result of the learners’ academic achievement was relevant with the learning strategies used and believed by them.38 _{220 of Thai graduate students} studying for Masters degrees at a public university in Thailand was participated to fill the questionnaire. It containing some questions about believe, language learning strategy used, and learning achievement. 35 of large group was chosen to be interviewed.

From that study, she was found that there was correlation between language learning strategy and academic achievement.39 _{In addition, the} believe of the strategies used by the learners also affected on their academic-achievement.

38 _{Kasma Suwanarak, “English Language Learning Beliefs, Learning Strategies and Achievement} of Masters Students in Thailand,” TESOL in Context Special Edition S3, 2012, 1.

24 CHAPTER III

RESEARCH METHOD

This chapter describes the features design of the fundamental research methods which were applied by the researcher. Inside it, there were any cases would be explained, they were research design, population and sample, research instruments, data collection technique, and data analysis technique.

A. Research Design

Research design was a procedure to collect, analyse, interpret, and report on data obtained during the research period. The study conducted by the researcher was classified as a correlational study, because the purposed of this study was to reveal the correlation between one variable to another variable. In this study, the researcher did not alter or modify variables, so this research was categorized as a survey research. The descriptive-quantitative method was used to analyse and report the result.

Correlational research was the method used to investigate variables (two or more) to determine the relationship diversity characteristic of variables. Another research asserted that correlational study means a statistical relationship between a set of variables, and none of them were manipulated.1 _{According to Creswell, correlational study was included in} the quantitative research.2 _{In addition, survey study was the part of}

1 _{David A. Kenny, Correlation and Causality (Storrs: Connecticut, 2004), 1.}

2 _{John W. Creswell, Research Design: Qualitative, Quantitative, and Mixed Method Approaches,} 2nd ed. (Thousand Oaks, Calif: Sage Publications, 2003), 13.

25

correlational study.3 _{It was designed to provide a quantitative or numeric} description of attitude, trend, or opinion of the population by studying a sample of it.4 _{Babbie (as cited by Creswell) stated that the survey research,} the procedure of the data collection was done by questionnaire or structured interview.5

According to the explanation above, the researcher investigated two variables in a condition where they were not manipulated. SES (Socio-Economic Status) and LLS (Language Learning Strategy) are two variables that would be investigated.

The first procedure was determining the object (population and sample). Because this research was universal (the results could be applied to any person), the sample could be selected randomly. The simple random sampling was used to determine the sample.

Furthermore, in the data retrieval from the sample was applied by using questionnaires. Questionnaire was a document that contains some questions or items used to investigate and collect the data would be analysed.6 _{According to Babbie (as cited by Heum Lee), A questionnaire}

could collect data by asking people with questions or asking them to agree

3 _{Ibid., 14.}

4 _{John W. Creswell, Research Design: Qualitative, Quantitative, and Mixed Method Approaches,} 4th ed. (California: Sage Publications, 2014), 145.

5 _{Creswell, Research Design, 14.}

6 _{Bidhan Acharya, “Questionnaire Design,” in A Paper Prepared for a Training Workshop in}

Research Methodology Organised by Centre for Post Graduate Studies Nepal Engineering College in Collaboration with University Grant Commission Nepal, Pulchok, June, 2010, 2,

accessed July 21,2016,http://www.saciwaters.org/CB/IFRM/IFRM/IV.%20Literature/Module%206_Qualitative

26

or disagree with statements representing different points of view.7 _The

questionnaire used in this study consists of two kinds, namely SES (Socio-Economic status) questionnaire and LLS (language learning strategies) questionnaires.

The third procedure was the process of data analysis. The results of the data collection were analysed using descriptive statistic and inferential statistic. In the descriptive statistic the normality test and homogeneity test was acted as the requirement before doing the inferential statistic. in the next procedure, MANOVA was carried out to analyse the hypothesis. Furthermore, Pearson product-moment formula used to determine the relationship between SES and LLS.

B. Population and Sample

1. Population

All item in any field of inquiry was defined as ‘Universe’ or ‘Population.8_{In this study, the population that was studied was the EFL or}

ESL learners in English Education Department, Faculty of Tarbiyah and Teacher Training, Sunan Ampel State Islamic University, Surabaya.. The total of the population was around 400 learners.

7 _{Sung Heum Lee, “Constructing Effective Questionnaires,” in Handbook of Human Performance}

Technology (Hoboken: Pfeiffer Wiley, 2006), 760.

8 _{C. R. Kothari, Research Methodology: Method and Technique, Second Revised. (New Delhi:} New Age International (P) Ltd, 2004), 55.

27

2. Sample

It was defined as the selected respondents was taken from the population.9 _{in the sampling technique, the researcher used random}

sampling methods. For the way of applying this sampling technique see the steps bellow.

75 learners of English Education Department on Faculty of Tarbiyah and Teacher Training in Sunan Ampel State Islamic University were participated as sample. 25 persons of each semester (start from 2nd _– 8th_{) voluntary was participate randomly. The total of the sample was taken} based on the theory of Gay and Diehl. They stated for the descriptive research (correlational or experimental) the total of sample should be 10% or more of the population total.10 _{This sample and population was taken by} considering these following reason

 Learners had enough experience and rated know about learning strategy they use than those who are still sitting in the 2nd _semester.  Learners had an awareness to know about their family or parents

economic and social background

 Students must have an English language background, because this research related to learning the English language.

 The learners in college had enough diverse culture to represent the diversity of SES.

9 _Ibid.

10 _{Edi Riadi, Metode Statistika Parametrik & Nonparametrik: Untuk Penelitian Ilmu-Ilmu Social}

28

C. Research Instrument

In this study, researcher used a single type of instrument in the form of questionnaire. Kothari said “A questionnaire consists of a number of questions printed or typed in a definite order on a form or set of forms.”11 The questionnaire was mailed to respondents who were expected to read and understand the questions and write down the reply. The respondents had to answer the questions on their own.

In this research, the researcher uses 2 type of questionnaires. The first questionnaire was purposed to measure the Language Learning Strategy used by students, and the second was a questionnaire that was served to investigate Socio-Economic Status owned by students.

1. Language Learning Strategies Instrument

For a questionnaire that measures LLS, the researcher used a questionnaire namely “Strategy Inventory of Language Learning for EFL/ESL v 0.7 (SILL)” which was designed by Rebecca Oxford (See appendix 1). The instrument consisted of 50 items based on taxonomy

developed by Rebecca Oxford. They were divided into Memory strategies, cognitive strategies, compensatory strategies, metacognitive strategies, affective strategies and social strategies. for the details explanation please read this following explanation:

a) Memory strategies (items 1–9) b) Cognitive strategies (items 10–23) c) Compensatory strategies (items 24–29)

29

d) Metacognitive strategies (items 30–38) e) Affective strategies (items 39–44) f) Social strategies (items 45–50)

The LLS questionnaire was categorized as international or standardised instrument, it was mean that the instrument validity and reliability was absolute, because it was often used as measurement tool without any change.

While for item scoring, every item had 1-5 score. The score was gotten from linier scale which was described as follow:

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

Every item had been answered got a score according the criterion above, then the result was calculated in each part of strategies of LLS. And the data was taken and proceed from the instrument was categorized as interval data.

2. Socio-Economic Status Instrument

The researcher used a questionnaire from a journal written by Aggarwal et al (See Appendix 2), because there were cultural

30

There are many questionnaires for measure SES, but they had no significant correlation with education, most of them use the SES questionnaires to measure the SES has correlation with health or economy. For education SES has several categories as the measurement tool, they are income, occupation, and education which were measured from family background aspect.

The total of question of SES questionnaire was 22 item. Every item was labelled as X (for example item number 1 was labelled as X1). For the data scoring, every answer of the data had been labelled with a number which was started from 1 and so on (it was depended on the total of answer items of multiple choices) Then the researcher calculated total of the score. It was the basic of SES classification.

According to Aggarwal the instrument still needed a test to check the validity and reliability. The researcher used pilot validity to check it. Sudarmanto asserted that the validity test could be applied by using Pearson Correlation Product Moment(PCPM) technique. the

researcher used IBM SPSS 23 to calculate the data. But in manual it was measured by this following formula.

= � ∑ � � �

�= ∑��= � �

�∑�_{�= �} − ∑�_�= ( _�) �∑_�=� _� − ∑�_�= ( _�)

a. Pilot Validity 1

The researcher distributed the questionnaire into 10 persons. After the data had been collected, the validity

31

investigated by using PCMP. Before performed it, the data was calculated to know about the total score.

After performing PCPM, the result of pilot validity 1

(see appendix 3 (Table 3.1. Pilot Validity 1)) showed there

were only 8 items of the questionnaire was decided as valid. The item could be valid if the r score (could be seen in correlation product moment column) > r table (could be seen on appendix 4 (r table)).

The significant value was decided by the researcher was 5% or 0,05. It was mean that the value of r table was 0,632. According to table 3.1, there only 8 item was decided as valid instrument, because they had a greater value than r table. they are:

1) Item x1=.817** > 0,632 2) Item x2=.683* >0,632 3) Item x3=.776* > 0,632 4) Item x4=.880** > 0,632, 5) Item x11=.726** > 0,632 6) tem x14=.734** > 0,632, 7) Item x19=.881** > 0,632 8) Item x20=.907* > 0,632

32

b. Pilot Validity 2

The fact was support the homogeny choice and the non-understandable factor was indicated as the reason about non validity of the items. So the researcher tried to give item 15 and 21 (see Appendix 5 (SES Instrument phase 2)) some

additional explanation/information/description based on the culture in Indonesia to make the items clear to understand and remove one item that hard to understand. so the total items were 21 from the previous item (22).

To solved the first problem (the answer had been chosen was homogeny), the researcher tried to add the total of the respondents. So in this Pilot Validity 2, the questionnaire was spread out to the 20 people which was different with the first pilot test. The increment of the respondent was purposed to make a more answer variety, so the data had a higher chance to be valid.

The method used to analyse the data was same with the

Pilot Validity 1. The result (see Appendix 6 (Table 3.2. Pilot Validity 2)) presented the same result as the previous test.

However, the significant score of the Pearson Correlation was different, but the valid item was same.

Because the total of respondents was 20, the r table was decided in sig. 5% was 0,444. From the table 3.2, the valid item

33

was still on the same item but in different value of correlation and significant (see the marking score in Appendix 6 (Table 3.2. Pilot Validity 2)). for more detail see the table 3.3 bellow

Table 3.3 (The conclusion of Pilot Validity 2) Number of

Items correlation (r Score of the score)

< (less than) or >

(more than) level 20 N r table on X1 X2 X3 X4 X11 X14 X19 X20 871** .687* .598* .889** .766** .843** .471** .867* more than more than more than more than more than more than more than more than 0,444 0,444 0,444 0,444 0,444 0,444 0,444 0,444

c. Pilot Validity 3

After holding the pilot validity twice, the researcher decided to take the valid item and used it as the instrument for this research. So the questionnaire of SES included 8 item (See

Appendix 7 (SES Instrument Used for Research)).

The chosen items were argued had capability to measure the variable, because the three main concept of Socio-Economic status (income, occupation, and education) had been included. For investigates why the data could be invalid, the researcher use and elicitation question which was related to each item, and the reason why the other item was clarified as invalid item was not because the respondents did not

34

understand about the question. But because they had their own reason such as follow:

1) They did not need it however they could have it (it was related with a question which ask about the proprietary, such as agricultural land, house, vehicle and other) 2) It was a privacy (it was related to the tax)

3) They do not know the fact (some of them did not know the real information of the family)

Before applied it to the real sample, the researcher spread out the chosen item to 10 persons, it was purposed to check the validity and their reliability. The reason why the researcher decrease the sample was the difficulties of the correlation test. one of the factor of it was the total of respondents. The higher of respondents’ total, the higher correlation probability could occur. If this test was passed with only 10 persons, it was mean that if it was applied to the respondents with higher mount, it will more valid.

The result of the Pilot Validity 3 (see Appendix8 (Table 3.4. Pilot Validity 3)) showed that in line Pearson Correlation indicated that all of item was clarified had a correlation each other. And the value of r score was higher than r table in 5% significant level (see table 3.4). So it could be concluded that

35

Table 3.5 (The conclusion of Pilot Validity 3)

Number of item Pearson

correlation score (r score)

< (less than) or >

(more than) significant level in R table in 5% 10 N X1 X2 X3 X4 X5 X6 X7 X8 0,897 0,669 0,776 0,706 0,880 0,874 0,737 0,670

> (more than) > (more than) > (more than) > (more than) > (more than) > (more than) > (more than) > (more than)

0,632 0,632 0,632 0,632 0,632 0,632 0,632 0,632

d. The Reliability Test

The validity test was run for three times and showed the same valid item, it was mean that the instrument was used by the researcher was reliable.

D. Data Collection Technique

According to Sudjana, data collecting could be applied by two styles, first was census and second was sampling.12 _{The technique used by} the researcher was a random sampling. 25 learners were chosen randomly from every semester in English education department. the researcher did not take the sample in single semester, because there was prohibition that the distribution to the population could not be effective, remembering that the population was EFL/ESL learners in English education department of Sunan Ampel Surabaya. By taking population from every semester, the distribution of the research result could be more effective. Sudarmanto,

36

asserted that in the random sampling, the sampling cloud be effectively when the population was bounded, then the sample could be taken.13

The research was carried out independently by the researcher. It needed time around 20 days to collect the data. The author spread out the questionnaire via online (by using social media, such as WhatsApp, BBM, and google form) and offline. But the much of information or data was arrived from offline method. 63 persons was filled the offline questionnaire, and 12 persons was filled the online questionnaire.

E. Data Analysis Technique

The data analysis technique contained about the way of the researcher to process the data had been collected. This section described about the statistical analysis used by the researcher. It was divided into two categorizations, first was descriptive statistic and the second was inferential statistic (parametric or non-parametric). The data analysis was performed by using IBM SPSS 23, so the descriptive statistic (included mean, median, mode, z score, and standard deviation) was calculated coincide with the inferential statistic.

This research used CPS (correlation product moment) to investigate the correlation between Socio-Economic status and language learning strategies. But before perform it, there were some requirements had to fulfil,14 _{they are:}

13 _{Sudarmanto, Statistik Terapan Berbasis Komputer: Dengan Program IBM SPSS 19, 47.}

14 _{Edi Riadi, Metode Statistika Parametrik & Non-Parametric (Tanggerang: Pustaka Mandiri,} 2014), 141.

37

1. The sample was random

2. The data was categorized as interval or ratio 3. The data had normal distribution characteristic 4. The data had homogeny characteristic

The 1st _{requirement had been completed because the sample was} taken using simple random sampling technique. Then the 2nd _requirement also had been fulfilled because the score of the data could be calculated. While mean, median, mode, standard deviation, and z score was needed to calculate before performing 3rd _{and 4}th _{requirements.}

The first step taken by the researcher carried out the descriptive statistic. In this step mean, median, mode, standard deviation, z score, and classifying the learners according SES level was applied.

The second step was applying inferential statistic. the researcher executed the normality test, homogeneity test, MANOVA test, and Pearson Correlation Product Moment test chronologically. For easier understanding see figure 3.1 bellow.

38

Figure 3.1 (MANOVA Steps) For further information, lets shift to the explanation bellow.

1. Descriptive Statistic

It was known as deductive statistic, simple statistic, and descriptive statistic. The purpose of it was to organize and analyse the numerical data, so it could give a description of the data clearly. Sugiyono (as cited by Sudarmanto,) asserted that the function of descriptive statistic was to describe the object of research by using sample or population without carried out an analysis.15

Mean, modus, median, standard deviation, table, graphic, and the others were some of the descriptive statistic. They were very important, because the inferential statistic was begun from them. The basic descriptive statistic was mean, median, and modus. After finding them the

15 _{Sudarmanto, Statistik Terapan Berbasis Komputer: Dengan Program IBM SPSS Statistik 19, 8.}

Step 1 (Descriptive Statistic)

•Calculate mean, median, and mode

•Calculate standard deviation

•Calculate z score

•Clasified learners according to the SES level

Step 2 (Inferential Satistic)

•Execute Normality test

•Execute Homogeny test

•Execute MANOVA

•Execute Pearson Correaltion Product Moment

39

standard deviation and z score could be found, then they could be used to carry out the normality and homogeneity test.

This was the first series of the data analysis way which must be calculated before applying the inferential statistic. In this part, the researcher had to found mean, median, mode, standard deviation, and z score. After calculated them, the researcher could classify the learners according to their SES level.

a. Mean

Mean could be defined as average score of the data. It could be calculated by using the formula bellow.

�= 1 + 2 + 3 +⋯. + �

� � = Mean

1− � = Data Frequency

� = Total of data/respondent

b. Median

Median was explained as the location of data which was disport it into 2 part with the same portion.16 _{It could be calculated} with:

� =�+ 1 2 Me = Median n = Total of data

40

c. Modus

While for the modus was explained as the value of the data with the most frequencies.17 _{To calculate the modus the} researcher only arranged the data form the smallest in to the highest. Then the score with highest frequencies could be found.

d. Standard Deviation (SD)

The next was standard deviation and z score. Standard deviation (SD) was used to measure how the data could be spread out. In general SD could be measured by using this formula.

=�Σ�� � −(Σ�� �)

�(� −1) = Variance

� = …… Score n = Sample Value

e. Z Score

After found the SD, another significant step was calculating the Z score. it could be used in almost parametric statistic. It also needed to measure the homogeneity and normality assumption test. In general Z score could be measure by this following formula.

41

= − �

��

x = the data would be calculated\ X = Mean

SD = Standard Deviation

f. SES classification

Table 3.1 (SES Classification Example)

Score reference Score of SES _Respondents Total of

LOW 1 – 17

MIDDLE 18 – 34

HIGH < 34

The researcher divided data into the 3 class, they are high, middle, and low. According to the table 3.1, the classification based on the range of the score had been divided. Learners who got total of score 1-17 was categorized as LOW SES, learners who got score around 18-34 was classified as MIDDLE SES, and they were get more than 34 of Total SES score was categorized as HIGH SES.

2. Inferential Statistic

After committed the descriptive statistic, and all of the component (mean, median, mode, z score, and SD) was found. The homogeny and normality test was carried out. Then if the data passed both of those assumption, MANOVA and Pearson Correlation Product

42

Moment could be used. For more detail information about the homogeny, normality, MANOVA, and Pearson Correlation Product Moment and how to interpret the result of them, let see this following information.

a. Normality Test/Normal Distribution Test (Kolmogorov

Smirnov Test)

After finding all of the requirement to apply the normality test and homogeneity test. the next step was calculating the normality test using Kolmogorov Smirnov method. Its application was equal with Lilliefors’ method, the differences was located on the comparator table. Lilliefors used the Lilliefors’ table and Kolmogorov Smirnov used Kolmogorov Smirnov’ table. Manually the Kolmogorov Smirnov test was carried out by these following steps.

1. Arrange the data from the smallest into the highest 2. Decide the frequencies of each data

3. Decide the cumulative proportion (CP) =

c a i ci

4. Decide the Z score for each data 5. Use the Z table

6. Use cumulative probabilities table (Fz)

7. Find out the value of a1 and a2 with this following criterion

43

a. _�2 = |�� − �_�| (the quarrel of CP and Fz) b. _�1 = |�2− ��

� |

8. Looking for the highest value of a1 and a2 (Dh) 9. See Kolmogorov Smirnov’s table (Dt)

10.Compare Dh and Dt. (Dh<Dt = Normal)

But the because the data will be analysed using IBM SPSS 23, the analysis was little differences. According to Sudarmanto, the result of Kolmogorov Smirnov test could be analysed by using the value of Asymp. Sig. (2-tailed) which was compared with Alpha Value.18 For the data interpretation, if the Asymp. Sig. had score more than Alpha score; it was mean that the data was normally distributed.

b. Homogeny Test (Levene’s Test)

While for the Homogeneity test was carried out by using Levene’s test. The homogeneity test had function to check was the data came from the homogeny population or not.19 _{Levene’s test was one of the method to carried out that} assumption test. In General, it could be calculated by using this following formula.

=( − �)Σ��(� − ��) (� −1)Σ(Vij−Vi)

18 _{Sudarmanto, Statistik Terapan Berbasis Komputer: Dengan Program IBM SPSS Statistik 19,} 130.

44

=Levene’s Value

�= Value of Residual Data = Mean

= Total of sample = Total of Population

To know the result of the data was homogeny or not, it could be seen from the significant value of each variable. When the score of sig. was more than 0,05 or 5% (Alpha), it could be concluded that the data was homogeny.

c. MANOVA

After both of the assumption was complete, the hypothesis test was carried out using MANOVA. It was equal with ANOVA, the differences was located on the independent variable.20 _{The researcher chose this method because the total} of variable was more than 2. If the variable only 2, it could be analyzed using T-test. While if the variable was 3, it could be analyzed by ANOVA. It was technique purposed to know about was the differences between variance score of one dependent variable was caused by the differences of each variable of independent variable.21 _{While for MANOVA the} total of dependent variable was more than two. According to

20 _{Jonathan Sarwono, Statistik Multivariat: Aplikasi Untuk Riset Skripsi (Yogyakarta: Andi} Yogyakarta, 2013), 19.

21 _{Abdul Muhid, Analisis Statistik: 5 Langkah Praktis Analisis Statistik Dengan SPSS For}

45

Sarwono, MANOVA was used to check the significant of mean between two groups with two dependent variable or more.22 The purposed of MANOVA was to investigate the hypothesis (by using MANOVA analyzation) and find out the differences of independent variable in dependent variable (by using Between Subject-Effect Test which had included in MANOVA).23

The analysis explanation of the first and second purposed of MANOVA was carried out by using the sig. value of each variable or variance. When the sig. value was less than Alpha score, it was mean that the null hypothesis was rejected.

d. Pearson Correlation Product Moment

It was purposed to investigate the kind of correlation between SES and LLS (it could be negative or positive correlation). The CPP (Correlation Product Moment) was used. It was defined as the test device was purposed to investigate the correlation between two random sample which was homogeny and normal.24 _{The analysis of data result was if the data had (*)} or (**) sign, it was mean that there was significant correlation.

22 _{Sarwono, Statistik Multivariat: Aplikasi Untuk Riset Skripsi, 19.} 23 _Ibid.

46

and the sign “- or +” in the score indicated the kind of correlation.25

25 _{Sami’an, “SPSS Korelasi” (presented at the Penggunaan SPSS Dalam Statistik, Universitas} Gadjah Mada, 2010), 10.

CHAPTER IV

FINDING AND DISCUSSION

This chapter presented about the result of the research. It was divided into 2 main aspects. The first was finding, it showed the result of the research had been proceeding. All the important discovery and the data processing was explained here, such as hypothesis test and the answer of the research including the requirement to apply (for example mean, median, modus, standard deviation, and z score). the second was discussion, it was purposed to answer the research objective and explore the research discovery and explain the limitation of it, then connected it with the previous study.

A. Finding

This section presented the important discovery of the research. The learners’ classification according to SES, normality assumption, homogeny assumption, MANOVA, and Pearson Correlation Product Moment were the topic was presented in this section.

1. The learners’ Classification According to Their SES Level

The data was gotten from the participant was divided into three groups based on the Socio-Economic point (see Appendix 9 (table 4.1. SES Classification)). It was printed that 27 persons were categorized

as learners with low SES, while 37 learners had middle SES family background, and the higher position of SES family level was owned by 11 learners. For the LLS the researcher did not show up the table

48

because there was not minimum or maximum score for LLS, it was depended on SES level, and it explained in the correlation product moment analyzation.

2. Normality Assumption

In order to check and investigate the normal distribution assumption, the Kolmogorov Smirnov test was used. The result (see

Appendix 10 (table 4.2 Normality Assumption)) showed that the

statistical significant (Asymp. Sig. (2-Tailed)) for Socio-Economic Status and All different language learning strategy showed indexes more than 5% or 0,05 (Asymp. Sig. > 0,05). For more simple see table 4.3.

Table 4.3 (The summarize of Kolmogorov Smirnov test)

Variable _{Sig. (as)} Asymp. Alpha _(A) Condition Conclusion

SES 0.054

0.05

(ss) more than (A) Normal

Memory 0.633 (ss) more than (A) Normal

Cognitive 0.114 (ss) more than (A) Normal

Compensatory 0.057 (ss) more than (A) Normal

Metacognitive 0.152 (ss) more than (A) Normal

Affective 0.274 (ss) more than (A) Normal

Social 0.197 (ss) more than (A) Normal

The SES, memory, cognitive, compensatory, metacognitive, affective, and social had Asymp. Sig. score more then 0,05, so it could be conclude that all of that variables was normal. It was mean that the

49

result of this researh was taken from the apportionment distribution of population, and it could be distributed in population.

The illustration of Socio-Economic status and language learning strategy distribution was shown in the normal curve, however the skewness indicate as little positive (see appendix 11 (figure 4.1 and 4.2)).

Figure 4.1 described the Socio-Economic status distribution.

the curve showed that the data of SES was normaly distributed, because it was located on the center, however the tailed was litle aslanted to right. It was mean that most of the sample was categorized in the midle of SES. But there was some learners have an extrim value of high SES, it could be seen from the long line in the right side.

Figure 4.2 described that average learners in English education

department had quite good language learning strategy, it was indicated from the top of the line. But there were some learners with high language learning strategies. It was indicated on the tail in the right side which was lower and longer.

3. Homogenety Assumption

It was one of requirement before performing MANOVA and Correlation Product Moment, because both of them was categorized of Parametric statistic. The function of homogeneity test was to check the equal data of variable (it was come from homogeny sample or not). Levene’s test was run to check it.

50

The result (see appendix 12 (Table 4.4 Test of Homogeneity of Variance)) point out that all of the significant value of mean had higher score than 0,05. For more detail see the following explanation.

Memory strategy = 0.158 (mean) > 0,05 = Homogeny Cognitive Strategy =0,719 (mean) > 0,05 = Homogeny Compensatory Strategy = 0,319 (mean) > 0,05 = Homogeny Metacognitive Strategy = 0,368 (mean) > 0,05 = Homogeny Affective Strategy = 0,576 (mean) > 0,05 = Homogeny Social Strategy = 0,368 (mean) > 0,05 = Homogeny The level of sig(_ߚ) of them were greater than 0.05 (the value of Alpha (_ߙ)), or it could be simplify as _ߚ> _ߙ. It could be concluded that every strategy of language strategy in each class of Socio-Economic status was homogeny.

The function of homogeny assumption was to minimize or it could prevent the error when the parametric statistic was applied. So when the data was analyzed by using parametric statistic, it could be more confidence and accurate, because the level of error was reduced or had been anticipated.

4. MANOVA (Multivariate of Analysis)

In MANOVA there was two test was carried out first was MANOVA and Between Subject-Effect test was the second.

CHAPTER V

CONCLUSION AND SUGGESTION

It was explained about the sum of the research according to the problems

which became the focus of this study. in addition, the researcher also gave

additional suggestion to the further research. The suggestion included the shortage

of the research had been carried out and the advices for the next research.

A. Conclusion

There was a positive linear correlation between Socio-Economic

status and language learning strategy. For more detail, every classes of

Socio-Economic background had their own level of language learning

strategies. The higher Socio-Economic status background owned by

learners, it means the better quality of language learning strategy used by

learners.

B. Suggestion

1. For further study, the researcher could investigate the effect of

Socio-Economic level on each strategy in language learning

2. In Socio-Economic status measurement, it was need more

investigation what was included in such as demand, want, need, and

others.

3. The important one of SES and LLS was the issues “how to solve or

create similar effective learning strategies between the gap of the SES

62

4. For teacher, it should be the sign to open the mind that not every

learner had equal condition, the teacher must do some research to find

the method that could be done by the all learners. The method has to

63

REFERENCES

Acharya, Bidhan. “Questionnaire Design.” In A Paper Prepared for a Training Workshop in Research Methodology Organised by Centre for Post Graduate Studies Nepal Engineering College in Collaboration with University Grant Commission Nepal, Pulchok, June, 2–11, 2010.

Accessed July 21, 2016. http://www.saciwaters.org/CB/IFRM/IFRM/IV.%20Literature/Module%2

06_Qualitative%20Research%20Methods/6.4%20Questionnaire%20Desig n_Acharya%20Bidhan.pdf.

Aggarwal, O. P., S. K. Bhasin, A. K. Sharma, P. Chhabra, K. Aggarwal, and O. P. Rajoura. “A New Instrument (Scale) for Measuring the Socio-Economic Status of a Family: Preliminary Study.” Indian Journal of Community Medicine 30, no. 4 (2005): 10–12.

———. “A New Instrument (Scale) for Measuring the Socio-Economic Status of a Family: Preliminary Study.” Indian Journal of Community Medicine 30, no. 4 (2005): 10–12.

Akhtar, Zarina. “Socio-Economic Status Factors Effecting the Students Achievement: A Predictive Study.” International Journal of Social Sciences And Education 2, no. 1 (2012): 281.

———. “Socio-Economic Status Factors Effecting the Students Achievement: A Predictive Study.” International Journal of Social Sciences and Education 2, no. 1 (January 2012): 281–287.

Australian Government. “Measuring The Socio-Economic Status of Higher Education Students.” 16. Department of Education, Employment and Workplace Relation, 2009.

Blevins, Brent M. Effects of Socio-Economic Status on Academic Performance in Missouri Public Schools. Lindenwood University, 2009. Accessed June 17, 2016. http://gradworks.umi.com/33/72/3372318.html.

Caro, Daniel H. “Family Socio-Economic Status and Inequality of Opportunity.” Freie Universität Berlin, 2010.

———. “Socio‐economic Status and Academic Achievement Trajectories from

Childhood to Adolescence.” CANADIAN JOURNAL OF EDUCATION 32, no. 3 (2009): 558–590.

Chamot, Anna Uhl. “Issues in Language Learning Strategy Research and Teaching.” Electronic journal of foreign language teaching 1, no. 1 (2004): 14–26.

64

Cowan, Charles D., Robert M. Hauser, Robert A. Kominski, Henry M. Levin, Samuel R. Lucas, Stephen L. Morgan, Margaret Beale Spencer, and Chris Chapman. “Improving the Measurement of Socio-Economic Status for the National Assessment of Educational Progress: A Theoritical Foundation.” National Center for Education Statistics, November 2012.

Creswell, John W. Research Design: Qualitative, Quantitative, and Mixed Method Approaches. 2nd ed. Thousand Oaks, Calif: Sage Publications, 2003.

———. Research Design: Qualitative, Quantitative, and Mixed Method Approaches. 4th ed. California: Sage Publications, 2014.

Dickinson, Emily R., and Jill L. Adelson. “Exploring the Limitations of Measures of Students’ Socio-Economic Status (SES).” Practical Assessment, Research & Evaluation 19, no. 1 (2014): 2.

Edi Riadi. Metode Statistika Parametrik & Non-Parametric. Tanggerang: Pustaka Mandiri, 2014.

Farooq, Muhammad Shahid, Azizul Haque Chaudhry, Mohammad Shafiq, and Girma Berhanu. “Factors Affecting Students’ Quality of Academic Performance: A Case of Secondary School Level.” Journal of quality and technology management 7, no. 2 (2011): 1–14.

Fewell, Norman. “Language Learning Strategies and English Language Proficiency: An Investigation of Japanese EFL University Students.” TESOL Journal 2 (June 2010): 159–174.

G.Carvalho, Renato, and Rosa F.Novo. “Family Socio-Economic Status and Student Adaptation to School Life: Looking Beyond Grades.” Kenneth N.Ross 10, no. 28 (2012): 1209–1222.

Griffiths, Carol, and A. I. S. St Helens. Language Learning Strategies: Theory and Research. AIS St Helens, Centre for Research in International Education, 2004.

Gürsoy, Esim. “Investigating Language Learning Strategies of EFL Children for the Development of a Taxonomy.” English Language Teaching 3, no. 3 (2010): 164.

Hansson, Åse, and Jan-Eric Gustafsson. Measurement Invariance of Socio-Economic Status across Migrational Background. University of Gothenburg, June 12, 2012.

Hirth, Marilyn A., and Richard C. Mitchell. “An Indiana Investigation of the Impact of Expenditures & Socio-Economic Status on Student Achievement: Does Money Matter?” Williamsburg: THE

65

EDUCATIONAL RESOURCES INFORMATION CENTER (ERIC), 1995.

Jennifer Barry. “The Effect of Socio-Economic Status On Academic Achievement.” Wichita State University, 2006.

Kenny, David A. Correlation and Causality. Storrs: Connecticut, 2004.

Kothari, C. R. Research Methodology: Method and Technique. Second Revised. New Delhi: New Age International (P) Ltd, 2004.

Liu, Hui-ju, and Chih-hui Chang. “A Study on Language Learning Strategy Use and Its Relation to Academic Self-Concept: The Case of EFL Students in Taiwan.” Journal of Language Teaching and Research 4, no. 2 (March 1,

2013). Accessed July 21, 2016. http://www.academypublication.com/issues/past/jltr/vol04/02/06.pdf.

Muhid, Abdul. Analisis Statistik: 5 Langkah Praktis Analisis Statistik Dengan SPSS For Windows. Surabaya: ZIFATAMA, 2012.

Okioga, Charles Kombo. “The Impact of Students’ Socio-Economic Background on Academic Performance in Universities, a Case of Students in Kisii University College.” American International Journal of Social Science 2, no. 2 (2013): 38–46.

O’Malley, J. Michael, and Anna Uhl Chamot. Learning Strategies in Second Language Acquisition. The Cambridge applied linguistics series.

Cambridge [England] ; New York: Cambridge University Press, 1990.

Oxford, Rebecca L. Language Learning Strategy: What Every Teacher Should Know. 1st ed. Boston: Heinle & Heinle Publisher, 1990.

Quagliata, Tiffany. “Ls There a Positive Correlation between Socio-Economic Status and Academic Achievement?” St. John Fisher College, 2008.

Rebecca L. Oxford. Language Learning Strategy. Boston: Heinle & Heinle, 2009. Riadi, Edi. Metode Statistika Parametrik & Nonparametrik: Untuk Penelitian

Ilmu-Ilmu Social Dan Pendidikan. Tanggerang: Pustaka Mandiri, 2014. Rothman, Sheldon. “The Changing Influence of Socio-Economic Status on

Student Achievement: Recent Evidence from Australia.” In The Changing Influence of Socio-Economic Status, 1–12. Melbourne, Australia: Australian Council for Educational Research (ACEReSearch), 2004.

Sami’an. “SPSS Korelasi” presented at the Penggunaan SPSS Dalam Statistik, Universitas Gadjah Mada, 2010.

66

Sarwono, Jonathan. Statistik Multivariat: Aplikasi Untuk Riset Skripsi. Yogyakarta: Andi Yogyakarta, 2013.

Sirin, Selcuk R. “Socio-Economic Status and Academic Achievement: A Meta-Analytic Review of Research.” Review of Educational Research 75, no. 3 (Fall 2005): 417–453.

Sudarmanto, R. Gunawan. Statistik Terapan Berbasis Komputer: Dengan Program IBM SPSS 19. Asli. Jakarta: Penerbit Mitra Wacana Media, 2013.

———. Statistik Terapan Berbasis Komputer: Dengan Program IBM SPSS Statistik 19. Jakarti: Mitra Wacana Media, 2013.

Sudjana. Metoda Statistika. 5th ed. Bandung: Tarsito, 1989.

Sung Heum Lee. “Constructing Effective Questionnaires.” In Handbook of Human Performance Technology. Hoboken: Pfeiffer Wiley, 2006.

Suwanarak, Kasma. “English Language Learning Beliefs, Learning Strategies and Achievement of Masters Students in Thailand.” 1–14. TESOL in Context Special Edition S3, 2012.

Valli Jayanthi, S., Santhi Balakrishnan, Angela Lim Siok Ching, Noor Aaqilah Abdul Latiff, and A.M.A. Nasirudeen. “Factors Contributing to Academic Performance of Students in a Tertiary Institution in Singapore.” American Journal of Educational Research 2, no. 9 (August 24, 2014): 752–758. Wilson, Ralph C. “Ls There a Positive Correlation between Socio-Economic

Status and Academic Achievement?” St. John Fisher College, 2008.

Z. Javid, Choudhary, Turki S. Al-thubaiti, and Awwadh Uthman. “Effects of English Language Proficiency on the Choice of Language Learning Strategies by Saudi English-Major Undergraduates.” English Language Teaching 6, no. 1 (December 11, 2012). Accessed July 21, 2016. http://www.ccsenet.org/journal/index.php/elt/article/view/23040.

Zare, Pezhman. “Language Learning Strategies among EFL/ESL Learners: A Review of Literature.” International Journal of Humanities and Social Science 2, no. 5 (2012): 162–169.