30 CHAPTER III
RESEARCH METHODOLOGY
This chapter describes a description of the research design, location and time of the research, focus of the research, population and sample, instrument of the research, procedure of collecting the data, and technique of data analysis.
A. Research Design
In this research, the researcher used the descriptive method. They are quantitative and approaches by using descriptive design. Abraham S. Fischer stated that “Quantitative research is a type of educational research in which the researcher decides what to study; asks specific, narrow questions; collects quantifiable data from participants; analyzes these numbers using statistics; and conducts the inquiry in an unbiased, objective manner.1
Besides, Margono stated that "Descriptive research is a research method that seeks to provide a systematic and careful with the facts and the nature of certain populations which aim to solve the actual problems faced now and collect data or information to be arranged, described, and analyzed".2
The research design of this research is a descriptive analysis. Descriptive analysis means the analysis of this research in the form of description, used statistically descriptive and also used Rasch Model Measurements in analyzing the data and information that has been collected.
1Abraham S. Fischler School of Education, Quantitative Research Methods, (NOVA:
Southeastern University). (Accessed on Oktober, 14 2020).
2Margono, Metode Penelitian Pendidikan, (Jakarta: Rineka Cipta, 2003), 2nd ed. P.158.
B. Location and Time of the Research 1. Location of the Research
The location of this research conducted at Madrasah Aliyah Darud Da’wah Wal Irsyad (MA DDI Kanang). It was located in Kanang, Jl. Mangondang No. 35 Desa Batetangnga, Kecamatan Binuang, Kabupaten Polewali Mandar,West Sulawesi.
2. Duration of the Research
The researcher used descriptive quantitative research to collect and analyze the data. Therefore, the researcher used more than one month for collecting the data from the students, teacher, and observation in school.
C. Focus of the Research
The focus of the research was to find out, analyze, and conclude data on students' difficulties. This research focuses on analyzes the students' difficulties in comprehending English text and factors causing students to faced difficulty in comprehending English text, which focused text in the form of explanation text.
The researcher focused on explanation text because it was based on the curriculum 2013, and syllabus of students at the second grade, and also it was recommendations from the English teacher, that second grade students of MA DDI Kanang still found it difficult in comprehending English text. That's why the researcher focused the study on analyzing the students’ difficulties in comprehending explanation text.
D. Participants and Sample of the Research 1. Participants
The participants of this research were the second grade students of MA DDI Kanang in the academic year 2020/2021 which consists of three classes. They were XI MIPA 1, XI MIPA 2, and XI IPS. The totals populations were 71 students. The distribution of the population was as follows:
Table 3.1 Population of Second grade MA DDI Kanang
NO CLASS/GRADE STUDENTS NUMBER
MALE FEMALE
1 XI MIPA 1 9 13 22
2 XI MIPA 2 8 12 20
3 XI IPS 18 11 29
Total 35 36 71
(Data Source: Administration Staff of MA DDI Kanang Academic Year 2020/2021)
2. Sample
A sample is defined as some members selected from the population.3 In this research, from the total population of the second grade students' of MA DDI Kanang, they are 71 students. The researcher used 41 students, as the sample of this research to obtain data about the difficulties faced by the students in comprehending English text, where the sample of this research retrieval used random sampling group techniques.
3Juliansyah Noor, Metodologi Penelitian Skripsi, Thesis, Disertasi & Karya Ilmiah, (Prenada Media).
This sample also was recommended from the English teacher that most of students of second grade were still low in comprehending English text, that's why the teacher recommended to the researcher to find and analyze difficulties that students faced and factors of causing the students’ still found difficult in comprehending English text, especially explanation text.
Table 3.2 Sample of Second grade MA DDI Kanang
NO CLASS/GRADE STUDENTS TOTAL THE
STUDENTS 1 SECOND GRADE
MALE FEMALE
12 29 41
(Data Source: Administration Staff of MA DDI Kanang Academic Year 2020/2021)
E. Instrument of the Research
To collect the data from the students, the researcher used to test and interview as the instrument of the research:
1. Test
After the researcher did the observations in the class, the researcher moved to the next steps of collecting data about the students' difficulties in comprehending English text used reading test. The test used was explanation text related to the research topic consist of 30 questions, the questions were multiple choice. The questions of the test were based on the five aspects of comprehending English text and each aspect will consist of 6 questions, so that would be 30 questions in the form of multiple choice. The test was used to find out the most difficult aspects that students faced in comprehending English text.
2. Interview
The interview occurs when the researcher asks the students as participants open-ended questions and record their answer. Then, the researcher transcribes and writes or types the data into a computer file for analysis.4 The researcher made an interview for students about the factors causing they found difficulties in comprehending English text, using factors difficulties from Westwood (Learner's background, Teaching technique, and Learner's environment), to get the data about the factors causing the students difficulties in comprehending. 12 questions were used to interviews the sample of this research.
F. Technique of Collecting Data
The technique of collecting data is a way to get data from the research fields.
This step is important to be presented so the researcher can do the research as well in the location of the research. The steps of data collection are presented to help the researcher in the process of collecting data. So, they can get information from the steps. The steps are explained as follows:
The researcher observed the situations and conditions from the school of MA DDI Kanang, especially for the second grade students of that school. The researcher interviewed the English teacher of the second grade to collect information about their students' ability and the problem faced by the students in comprehending.
Because the school was on holiday, so the researcher gathered several students at the school mosque, to answer the reading test questions that were used
4John W. Creswell, Educational Research: Planning Conducting and Evaluating Quantitative and Qualitative Research, Fourth Edition (USA: Pearson Education, 2008), p.217.
by the researcher to identify the students' difficulties in comprehending explanation text. And also conduct interviews with several students to determine the factors that caused students to experience difficulty in comprehending English text and recorded using a telephone.
G. Technique of Data Analysis 1. The Analysis of Reading Test
Data analysis in this study used the Rasch Model data analysis technique.
According to Bond and Fox, this model is a one-parameter item response theory (IRT), model. That presupposes that each item is a difficult parameter. This model also arranges abilities and items based on difficulty. The Rasch Model also converts raw data to interval data with the same value between units and other units.
The basic principle underlying the Rasch Model is the probability of the respondent to answer any item correctly based on the difficulty of the item and the respondent's ability. The following is an analysis of the Rasch Model.
a. Item and Respondent Reliability (Item and Person Reliability)
According to Linacre, the item reliability index means that the difficulty of the item is the same as other samples that have the same ability. Meanwhile, the reliability index of the respondents means that the individual abilities in this sample are consistent even though they are given other instruments that measure the same construct. The maximum reliability index value is one. This value indicates high reliability. The item and respondent reliability index received was ≥ 0.80 (Bond and Fox, 2015). This is following the opinion of Fischer (2007) that the item reliability index and respondents who were well
received were > 0.80. As opposed Fischer (2007) divides the reliability criteria, namely <0.67 (weak), 0.67-0.80 (enough), 0.81-0.90 (good), 0.91-0.94 (very good) and > 0.94 (special).5
b. Stratum Alienation Index (Item Separation)
The item separation index showed the isolation for the item difficulty stage and the respondents’ isolation index (Smith & Smith 2004).6 The index of isolation of the respondents shows the number of ability strata that are known in the sample group. The value of the index of isolation to explain the ability of the items and respondents received, according to Bond & Fox (2015); Linecre (2002) is a value ≥ 2. Fischer (2007),7 divides the isolation of the respondent’s stratum and items, namely 2 or <2 (weak), 2-3 (enough), 3-4 (good), 4-5 (very good), > 5 (special).
c. Polarity Item or (Point Measure Correlation)
Polarity analysis or item suitability is an indicator used to move in one direction intended by the construct by the measured. Polarity analysis uses the point measure correlation (PMC) analysis technique or points measure correlation, which is a technique to produce items that match the measured conduct. The PMC 1.0 value indicates that all respondents with low ability answered the item incorrectly and all respondents with high ability answered the item correctly. The index of PMC values received is between 0.4 and 0.85
5Wahyu Hidayat, Sri Mulianah, Mujahidah, Analysis of the National Character Senior High School Students by Using Rasch Model, (Indonesia: INCRE, 2019).
6Smith & Smith, Introduction to Rasch Measurement, (Maple Grove, MN: JAM, 2004).
7Fischer, RATING Scale Instrument Quality Criteria, (Rasch Measurement Transactions, 2007).
(Azrilah et al.). Meanwhile, Alagumalai et al. categorized PMC values into five, namely: > 0.40 (special), 0.3 – 0.39 (good), 0.20 – 0.29 (sufficient), 0.00 – 0.19 unable to distinguish, and <0.00 (need-to recheck items).8
d. Item Fit Order
Examination of fit items was carried out employing infit mean square (MNSQ) analysis. MNSQ ideal value is one. To see the item suitability criteria, it can be determined from the Mean Square outfit index (Azrilah et al. 2013;
Bambang and Widhiarsono 2014; Bone et al. 2014).9 The MNSQ output values received are shown in the following table.
8Alagumalai, dkk, Applied Rasch Measurement: Book of Examplars, (Papers in honour of John P. Keeves. Dordrecht: Springer, 2005).
9Azrilah, dkk, Asas Model Pengukuran Rasch: Pembentukan Skala dan Struktur Pengukuran, (Bangi: Universiti Kebangsaan Malaysia, 2013).
Table 3.3 Criteria of Accepted MNSQ Outfit
MNSQ Keterangan
Wright, B. D., & Linacre, J. M
(1994) 0.8 - 2.0 MCQ (High Stakes)
0.7 - 1.3 MCQ (run of the mill) 0.6 - 1.4 Rating Scale (Survey) 0.5 - 1.7 Clinical (Observation)
0.4 - 1.2 Judged (agreement encourage) Borg & Fox (2015) 0.7 - 1.3 Model fit/good fit
< 0.7 M isfit very low/less model/overfit
> + 1.30 M isfit very large/under fit Fischer, W.P Jr (20007) < 0.33 (Weak)
0.34 - 2.9 (Enough) 0.50 - 2.0 (Good) 0.70 - 1.4 (Very Good)
0.77 - 1.3 (Special)
This research uses political data (rating scale), so the researcher follows Linecre's (2002) opinion where the accepted MNSQ outfit interval is 0.6 – 1.4.
e. Impressing a Single Construct (Dimensionality Item)
Unidimensional assumes that the items on the test measure a single ability (Alagumalai et al. 2005).10 This model provides a concise procedure for checking the dimensions of the measurement results. Unidimensional measurements can be done with PCA (Principal Component Analysis). The following table shows the dimensionality following the criteria.
10Alagumalai, dkk, Applied Rasch Measurement: Book of Examplars, (Papers in honour of John P. Keeves. Dordrecht: Springer, 2005).
Table 3.4 Dimensionality Follow Criteria
Criteria Weak Enough Good Very
Good Special Variants in data described
through measurements in data <40% 50 - 60% 60 -
70% 70 - 80% >80%
Unexplained variants according
to contras 1 Residual PCA >15% 10 -15% 5 -
10% 3-5% <3%
f. Differential Item Functioning (DIF)
To find out differences in the ability of respondents in answering items based on gender, religion, or other variables, it can be done by analyzing differences in item functions based on DIF values (Paniagua 2005; G Hays 2013). According to Bond and Fox,11 significant DIF is based on three indicator values, namely the t value, the DIF contrast value, and the p-value. The measure of the t value lies at + 2.0 or t ≥ ± 2.0, the DIF contrast index is at the value of ± 0.5 ≤ -0.5 and the significant value for p is p <0.05.
2. The Analysis of Interview
In analyzing the resulting interview from the students' answers, the researcher analyzes the data used three steps: they are data reduction, data display, and conclusion or verification.12 The procedure in analyzing the students answer from the interview that has been done by the researcher will describe as follows:
11Bond & Fox, Applying the Rasch Model: Fundamental Measurements in the Human Sciences, (New York: Routledge, 2015).
12Setiadi, Ag, Metode Penelitian untuk Pengajaran Bahasa Asing: Pendekatan Kuantitatif and Kualitatif, (Yogyakarta: Graha Ilmu, 2006).
a. Data Reduction
The first step is to make the analysis using data reduction. Summarize the data chose the basic thing, focus on the important one, and sough themes and patterns/ this data have been reduced. In reducing the data, the researcher is guided by the objectives to be achieved: the main objective of the qualitative research is the finding. Therefore if in the process of data reduction the researcher found unknown patterns, that was precisely what should be the concern of the researcher in performing data reduction because it is the process of thinking that requires sensitive intelligence, flexibility height, and depth of insight.
b. Data Display
In data display, the researcher is presenting the data that has been reducing before. In quantitative research, the presentation of the data will use a table, graphs, and so on. However, in qualitative research, the data will be present uses the form of description, among categories and relationships. And will be explained in the form of narration.
c. Conclusion or Verification
The last step in analyzing the answer from the students' interview is conclusion or verification. After the researcher reduces and displays the answer of the students' interview then conclude the result of both steps. This last step is used to explain the result of the procedure in analyze descriptive qualitative research.