Page 992 - MARSIUM'21 COMP OF PAPER
P. 992
3.0 RESEARCH METHODOLOGY
3.1 RESEARCH DESIGN
The study was designed the conceptual blueprint of the researcher, including all sources, methods and techniques used for collecting data,
measuring data and analyzing data to achieve research goals (Akhtar, 2016). Ahuja (2010) stated that research design is set of conditions to
collect and analyze data in an integrated manner of economic and procedural relevance to research purposes. There are two types of research
designs: exploratory design and conclusive design. Malhotra and Dash (2016) explained that exploratory design method is used to gain
insights and comprehend the nature of the marketing phenomenon. However, conclusive design is to examine specific hypotheses and
investigate relationships among variables. In exploratory research, qualitativedata is used, and its research process is relatively less structured
and flexible. On the other hand, the conclusive design research process is structured and formal. There are two types of conclusive design
research: descriptive research and causal research. Descriptive research is to depict the attributes of related groups, such as customer,
salespeople and organizations. On the other hand, causal research is to examine the relationships between independent variables and
dependent variables of a phenomenon (Malhotra, 2011). Conclusive design is to test the hypotheses and investigate relationships. The
research procedure is formal and well-structured. In this study, the conclusive research design is used to investigate the relationship between
social media interaction, customer brand engagement, brand awareness and brand image of a pastry shop. From the findings, social media
brand engagement strategies are implemented to raise the brand awareness and brand image of Chop Kek Lian Seng.
3.2 POPULATION AND SAMPLING
According to Krieger (2012), a population is defined as the total number of people or units being studied in research. In addition, the sampling
refers to a list of actual cases from which samples will be taken. The target population of this study is individuals who are using social media
platforms as interaction with others in Malaysia to buy pastry products. The non-probability sampling methodology which is the purposive
sampling technique is used for data collected. According to Suhr (2006), sample-to-item ratio could be determined sample size of the study
based on the numbers of items and it must not less than 5 to 1 ratio. This study consists of 33 items(questions) would require 165 of
respondents.
3.3 RESEARCH INSTRUMENT
A research instrument is a tool used to collect, measure and analyze data from the research participants. The research instrument depends on
the type of study being conducted; quantitative, qualitative or combination of the two (Taherdoost, 2016). The quantitative approach is used
in this study, and questionnaires is used to collect data. The questionnaire is divided into 2 sections: Section A and Section B. Section A
contain 10 questions about respondents' demographic profiles such as the age, gender, countries of origin, and frequency of using social
media platforms. Section B consists of 26 questions that are related to brand awareness, brand image and social media interaction. All the
measurement item adapted from previous study by from Cheung et al. (2020). There are 4 items to measure social media interaction, 10 items
for customer brand engagement, 5 items for brand awareness and 7 items for brand image. In addition, the questio nnaire featured 5- point
Likert scales 1 (strongly disagree), 2 (disagree), 3 (neutral), 4 (agree) and 5 (strongly agree). After that, the data is screened, processed and
analysed with the Statistical Package for the Social Sciences (SPSS).
4.0 DATA ANALYSIS
4.1 RESPONSE RATE
Quantitative method was used by this study to distribute surveys to respondents via Google form. The minimum sample size of this study is
165. A total of 225 sets of questionnaires was distributed to the targeted respondents. However, there are 50 sets of returned unusable
questionnaires due to the respondents did not like and follow any traditional pastry shop in social media. Therefore, only 175 questionnaires
were usable with response rate of 78%.
4.2 OUTLIER
The technique of discovering outliers, or anomalous observations in a dataset is known as outlier analysis (Bush, 2020). Therefore, in this
study Statistical Package for the Social Sciences (SPSS) was used to detect the incredibly small or incredibly large values of the dataset. If
an extreme value is discovered, it may be due to a data error. Therefore, it must be removed from the dataset (Denis, 2019). The results found
that customer brand engagement construct had outliers for samples of 127, 173, 104 and 151. On the other hand, brand awareness construct
had outliers for samples of 10, 127, 173, 74. Meanwhile, brand image construct had found an outlier for sample 74. In total, there are 6
samples should be deleted and 169 dataset was used for further data analysis.
993

