20
CHAPTER III
RESEARCH METHODS
3.1 Research Design
The research approach used is qualitative with a descriptive research design. This
approach was chosen because the researcher aims to describe and provide an overview
of students' perceptions regarding the use of YouTube as a learning media for English
speaking among eleventh-grade students at SMA Muhammadiyah Karang Intan. In this
research, the researcher analyzes the data to describe students' perceptions regarding the
use of YouTube in learning speaking skills among eleventh-grade students at SMA
Muhammadiyah Karang Intan.
1.2 Subject and Location of the Research
3.2.1 Subject of the Research
The subjects of this research are eleventh grade students at SMA Muhammadiyah
Karang Intan. The selection of the eleventh grade as the research subjects was based on
the consideration that students at this level have a stronger foundation of English
knowledge compared to students in previous grades, making them more prepared to
face more challenging learning methods, such as the use of digital media in the teaching
and learning process. Furthermore, eleventh-grade students are in a crucial phase where
they begin preparing for their final school exams, making English-speaking skills
increasingly important to support their academic abilities.
3.2.2 Location of the Research
This research was conducted at SMA Muhammadiyah Karang Intan, located at
Jalan Melati, Pasar Baru No.63, RT.03, Kecamatan Karang Intan, Kabupaten Banjar,
Provinsi Kalimantan Selatan. The selection of this location was based on the fact that
the school is situated in a rural area, which generally receives less attention in terms of
educational resources compared to urban schools. Additionally, SMA Muhammadiyah
20
21
Karang Intan is a private school managed by a community organization, which presents
unique characteristics and challenges in terms of management and provision of educational
facilities. Despite these challenges, the school has provided adequate facilities to support
teaching and learning activities, including in the teaching of English.
The selection of eleventh-grade students was also made after a brief interview with the
students and the English teacher at the school. The teacher involved has given permission for
this research and revealed that they have previously used YouTube as a learning medium to
teach speaking skills. This indicates that the students already have prior experience with
using YouTube in English learning, allowing them to provide deeper and more relevant
perceptions regarding the medium.
Therefore, both the location and the research subjects can be used to explore “Students'
perceptions of the use of YouTube as a learning medium for English-speaking skills among
eleventh-grade students at SMA Muhammadiyah Karang Intan”.
3.3 Source of the Data
Data collection methods in this study involve the use of a questionnaire and interviews.
The questionnaire is employed to gather responses from students regarding their perceptions
of using YouTube as a learning medium for English speaking skills among eleventh-grade
students at SMA Muhammadiyah Karang Intan. Interviews are conducted to gain deeper
insights into students' perceptions and their experiences using YouTube for learning
purposes.
3.4 Data Collection Techniques
In the process of data collection, the researcher used two primary methods: questionnaires
and interviews. Both methods were employed to investigate students' perceptions of
YouTube as a learning medium in developing speaking skills. Below is a detailed explanation
of the data collection techniques used in this study:
1. Questionnaire
The questionnaire is a research instrument consisting of a series of questions or statements
to be answered by respondents. It was used to collect information related to facts, opinions, or
attitudes. This study employed closed-ended questions, in which the researcher presented
statements with predefined answer choices based on a Likert scale.
Each item in the questionnaire was designed to measure various aspects relevant to the
objectives of the study. As Creswell (2014) states, the Likert scale is an effective tool for
21
22
measuring students’ perceptions and attitudes in educational research. The Likert scale used
in this research consists of four response options:
1 = Strongly Disagree
2 = Disagree
3 = Agree
4 = Strongly Agree
The questionnaire in this study was adapted from a previous study by Fitri Damayanti and
Desi Wijayanti Ma’rufah entitled "Students' Perception on the Use of YouTube as Learning
Media in Speaking at the 9th Grade of MTs NU 01 Kroya." The items and statements in the
questionnaire were modified to suit the specific needs of this study.
The following table illustrates the theoretical framework, indicators, and the related
questionnaire items:
Theory Indicators Items
Yuhariah & Syafryadin (2023); Rorimpandey
The Function of YouTube as
(2019); Anggrarini & Faturokhman (2021); Q1, Q2, Q3
Learning Media
Maziriri et al. (2020)
Jallaludin (2016); Irwanto (2002); Sahin The Advantages of YouTube as a
Q5, Q6, Q10
(2016) Learning Media
Jalaluddin (2016); Vandergrift & Goh The Advantages of YouTube in Q11, Q12, Q13, Q16–
(2012); Harmer (2007); Mulia et al. (2020) Developing Speaking Skills Q20
Jallaludin (2016); Irwanto (2002); Sahin The Disadvantages of YouTube as
Q4, Q7, Q8, Q9
(2016) Learning Media
Asworo (2019); Kusumawardani &
The Disadvantages of YouTube for
Mardiyani (2018); Mutia et al. (2020); Q14, Q15
Developing Speaking Skills
Haidara (2016)
The questionnaire scores were not analyzed using deep quantitative methods. Instead, they
were used to classify students’ levels of understanding into three categories: high, medium,
and low. This categorization aimed to select interview participants from various levels of
understanding to gain more comprehensive and diverse insights.
The classification of students' understanding levels was determined based on their
average scores from the questionnaire responses. The categorization was as follows:
High: average score ≥ 3.3
Medium: 2.7 ≤ average score < 3.3
Low: average score < 2.7
22
23
This classification approach aligns with the guidelines provided by Widoyoko (2012),
who suggests that Likert scale scores can be categorized into qualitative levels to facilitate
interpretation in educational research contexts.
The questionnaire consisted of four main parts. The first part collected demographic
information such as name, age, class, and WhatsApp number. The second part focused on the
purpose of the research. The third part presented an overview related to the research topic.
The fourth part contained specific statements used to collect the required data.
A total of 31 students participated in filling out the questionnaire, which was distributed
directly in the classroom setting.
2. Interview
The interview served as a supplementary method to explore deeper insights based on the
questionnaire results. Through interviews, the researcher could investigate participants’
perspectives, ideas, beliefs, and feelings about the topic in their own words.
A semi-structured interview format was used, in which the researcher prepared a set of
guiding questions while allowing open responses from participants for richer exploration. The
interviews were recorded using a phone and important points were also noted.
Some of the interview questions were derived and modified from the questionnaire to
ensure coherence between the two instruments. The research was conducted in two stages. In
the first stage, the questionnaire was distributed and preliminary data were collected and
analyzed. In the second stage, semi-structured interviews were conducted with participants
selected based on the initial questionnaire results. Data from the interviews were then
analyzed and combined with questionnaire results to achieve a deeper and more
comprehensive understanding of the research topic.
This study employed purposive sampling to select interview participants. As Creswell
(2012) explains, purposive sampling is a technique in which participants are deliberately
chosen based on specific characteristics relevant to the objectives of the study. In this
research, those characteristics were determined based on the average questionnaire scores.
To further enrich the findings, the researcher also applied the maximum variation
sampling strategy. According to Patton (2002), this strategy aims to capture a wide range of
perspectives by selecting participants with significantly different experiences or views.
Therefore, participants were chosen from three different categories—high, medium, and low
—based on their questionnaire results. This approach was intended to identify common
patterns emerging across various student experiences.
23
24
3.5 Technique of Analyzing the Data
Qualitative data analysis involves interpreting text and images to answer research
questions. This process systematically organizes and examines interview transcripts, field
notes, and other materials to deepen understanding and effectively present findings to others.
In this study, the researcher used the Miles and Huberman (1994) model for data analysis:
a. Data Reduction
Data condensation involves selecting, focusing, simplifying, or abstracting data from
the analysis of questionnaires, interview transcripts, and documents. This process aims to
strengthen data that is relevant to the research questions. Data on students' perceptions of
using YouTube as a learning medium for speaking skills were collected through
questionnaires and interviews, then transcribed. Irrelevant data were excluded. After
collecting and condensing the data, the researcher presented it descriptively.
b. Data Display
After the data is collected and simplified through the data reduction process, the next
step is to present the data in an organized manner. Data display is the process of
systematically organizing the collected information so that the researcher can draw
conclusions and make decisions based on the data. Data display is a crucial step because it
helps the researcher and readers to understand the data more clearly and comprehensively.
In qualitative research, the data often consists of long narratives from interviews,
observations, or field notes. To make it easier to understand, this data needs to be
simplified without losing its essence and arranged in a form that allows conclusions to be
drawn. One common way to present qualitative data is through structured descriptive text,
where information is organized into clear and logical paragraphs.
c. Drawing Conclusions
The final stage of analysis is drawing conclusions and verification. Conclusion
drawing is the stage where the researcher begins to formulate understanding or
generalizations from the data that has been analyzed. In qualitative research, conclusions
are often not drawn all at once at the end but through a continuous process of reflection.
As the researcher gathers more data, they may find that initial conclusions need to be
adjusted or revised. Verification is the process of ensuring that the conclusions drawn are
truly accurate and supported by the existing data. This may involve re-checking the data,
comparing results from various sources, or gathering additional data to support or refute
the conclusions drawn. The final conclusions may only be finalized after all relevant data
24
25
has been collected and verified. Thus, drawing conclusions in qualitative research is a
dynamic process, where initial conclusions may change as more data is collected.
Conclusions are verified to ensure their validity, and only after all necessary data has been
collected and analyzed can the final conclusions be finalized.
3.6 Validity of the Data
Triangulation is an approach that intentionally uses more than one method of data
collection and analysis to study social phenomena. The purpose of this technique is to
examine the consistency of findings obtained through various data collection methods.
According to Walidin et al. (2015), there are four types of triangulation techniques:
1. Data Triangulation
Data triangulation involves using various data sources to verify specific information.
For example, if researchers want to understand students' perceptions of using YouTube as
a learning medium, they can collect data through questionnaires and interviews. By
comparing the results from these different sources, researchers can ensure that the
information collected is accurate and consistent.
2. Investigator Triangulation
Investigator triangulation involves more than one researcher in the data collection and
analysis process. This aims to reduce individual bias that might occur if only one
researcher is involved. By involving multiple researchers, interpretations of the data can
be compared to find similarities or differences in their views.
3. Methodological Triangulation
Methodological triangulation involves using various methods to collect data and
comparing the results from different methods. This helps ensure that the research findings
are not limited to a specific context or method.
2. Theoretical Triangulation
Theoretical triangulation involves comparing research findings with existing theories
or literature to avoid bias from individual researchers. By examining how the findings
align with various theoretical perspectives, researchers can ensure that their conclusions
are based on a broader understanding and not just on their own assumptions.
Through triangulation, it is expected that by utilizing various perspectives or
viewpoints, the results obtained will be closer to the truth. In this study, among the four
triangulation techniques described, the researcher chose to validate the data using the first
technique, which is to explore the truth of specific information using various methods and
25
26
data sources. The researcher compared the data obtained from questionnaires and
interviews.
26