CHAPTER 3METHODOLOGY3.1 IntroductionThis chapter explains the method adopted by this research, describing the component involved in conducting this research from the population, population frame and sampling techniques used for the questionnaire. Finally, this chapter provides a detail explanation of the selected mode of analysis used and the data collection method.3.2 Research DesignGenerally, research design means a structure to plan and execute particular research. Research design is the hardest section of the research as it includes all the four important considerations such as the strategy, the conceptual framework, the identification of whom and what to study on and the tools and procedures to be used for collecting and analysing data.
The research design basically is divided into several types, for example, qualitative research and quantitative research (Holmes R. et al, 2005). Therefore, in order to satisfy the objectives of the study, quantitative research was used by the researcher in this study. Quantitative data sources include surveys, questionnaires, data collection, observations, and data analysis.
The advantage of quantitative research is that it offers a complete description and analysis of a research subject and help the researcher to measure and analyse the data to provide insight about the findings of the research (Collis & Hussey, 2004).3.3 Sources of dataData collection is done through both secondary and primary sources (Salkind, 2007). In this study, primary data is used. A primary data source is an original data source, that is, one in which the data are collected by the researcher for a specific research purpose or project (Salkind, 2007). Primary data collection is quite expensive and time consuming compared to secondary data collection. The primary data used in this study is questionnaire on the students’ perceived stress and their locus of control. The researcher used questionnaire in order to conduct the research analysis. The primary data that have been collected is very useful to cover every aspect of the study. 3.4 Methodology used in the study An extensive literature review was conducted to gather information on research objectives such as factors on perceived stress, physiological problems, and locus of control. The researcher gathered literature studies and schedule to get more information in this study. The materials are from the journal, texts book and research papers gathered from libraries and the Internet. The objectives of the research were to identify the main cause of stress related to the 6 domains to analyse which domains affect the students’ physiological problems. Besides, this study also aims to gather respondents’ insights on what are their locus of control whether internal or external and how their internal and external locus of control affects their level of stress (e.g. high stress or low stress).For the purposes of this research, questionnaires were used. The questionnaire is a set of questions has been prepared to ask a number of questions and collect answers from respondents relating to the research topic. A questionnaire is a series of questions asked to individuals to obtain statistically useful information about a given topic. When properly constructed and responsibly administered, questionnaires become a vital instrument by which statements can be made about specific groups or people or entire populations. As far as data collection tools were concerned, a set of questions was formulated to gain insight and opinion on the selection of research characteristics. The questionnaire has been constructed and data were collected by UM undergraduate students from Faculty of Economics and Administration on their project for the course EIA1006 in year 2018. The researcher used the provided data to further analyse and provide insights on the findings. The questionnaires only adopted close-ended questions in the survey that was divided into four sections. Section A aimed to find the stress-related causes on 6 domains where the items in this section are expressed as a 5-point Likert scale from no stress to extreme stress whereas Section B aimed to find out their aspects and symptoms of psychological stress where the items are a yes or no. Besides that, Section C is aimed to find out the locus of control among UM undergraduate students and Section D consisted of general questions to identify the profile of respondents. The survey questionnaire is included in Appendix 1 (see Figure 1). There were 260 respondents obtained from the data collected by the students of EIA1006 in year 2018. For the purpose of this study, it was important to identify respondents’ level of stress on the 6 domains and their psychological stress. Further, the study is also on their locus of control to identify whether the respondents are more to an internal or external locus of control and to analyse the level of stress of the students with external and internal locus of control (e.g. high stress or low stress). The results will be provided and discussed in the following sections.The method of purposive sampling was used to develop the sample of the research under discussion. According to this method, which belongs to the category of non-probability sampling techniques, sample members are selected on the basis of their knowledge, relationships, and expertise regarding a research subject (Freedman et al., 2005). From the obtained data, the sample members who were selected are only undergraduate students at University of Malaya because the study is within the University of Malaya. There were 260 respondents who answered the questionnaire and the students EIA 1006 in year 2018 used convenience sampling to carry out the research. 3.5 Analytical MethodsData analysis is a very important segment in the research. Data analysis gives the ability to researchers to structure the qualitative data collected in a way that satisfies the accomplishment of research objectives. However, human error is highly involved in content analysis, since there is a risk for researchers to misinterpret the data gathered, thereby generating false and unreliable conclusions (Krippendorff & Bock, 2009). Percentage and frequency values are utilized to analyse the profile of respondents. In order to identify the cause of stress related to domains, first, the researcher carries out the evidence for content validity. The researcher used the factor analysis and found that there are 6 domains in the study which affects stress level. Exploratory factor analysis (EFA) is a type of factor analysis that is utilized to identify the number of latent variables that underlies an entire set of items. This analytic method needs to be considered exploratory because the links between the indicators and variables (factors) are uncertain (Coakes and Steed, 2003). Since in this study, most of the items have been developed from prior studies, it is very important to make sure that they are properly connected to their basic factors (variables). To achieve this purpose, the utilization of exploratory factor analysis is deemed necessary in this study. Exploratory factor analysis detects the constructs. For example, the factors that underlie a dataset based on the correlations between variables which means the questionnaire items (Field, 2009). The factors that explain the highest proportion of variance the variables share are expected to represent the underlying constructs. In contrast to the commonly used principal component analysis, factor analysis does not have the presumption that all variance within a dataset is shared (Costello & Osborne, 2005). Since that generally is not the case either, factor analysis is assumed to be a more reliable questionnaire evaluation method than principal component analysis (Costello & Osborne, 2005). Factor loading specifies the strength of the relationship between the item and the latent variable and thus, is utilized to determine the convergent and discriminate validity of the scales (Hair et al., 2006). The dimensions of the scale were examined by factor analysing the items using the principal components analysis with Varimax rotation. Minimum eigenvalues of 1.0 helped determine the number of factors or dimensions for each scale (Hair et al., 2006). Factor loadings greater than 0.50 are generally needed for practical significance (Fen and Sabaruddin, 2008), where the items for a factor are retained only when the absolute size of their factor loading is above 0.50 (Fen and Sabaruddin, 2008). In Chapter 4 (table 2) shows the results of the factor analysis of the variables. Internal reliability of measurement is utilized in multi-item scales, and it turns to its consistency (Hair et al., 2006). Internal reliability refers to whether those items are internally consistent or whether the items that constitute the scale are measuring a single concept (Hair et al., 2006). Estimates of reliability based on the average correlation among items within a test, concern internal consistency (Lee, 2001). The internal consistency will be high if the correlation offers a high result (Lee, 2001). Nunnally (1978) suggested that an acceptable level of coefficient alpha to retain an item in a scale is at least 0.70 score. The researcher used the Cronbach’s alpha reliability to measure that determines the degree to which an internal consistency exists among the indicators within the same variable or factor coefficient (Lu et al., 2007). It normally ranges between 0 and 1. However, there is actually no lower limit to the coefficient. The closer Cronbach’s alpha coefficient is to 1.0 the greater the internal consistency of the items in the scale (Joseph and Rosemary, 2003).The items are based on a 5-point Likert scale in the data gathering tool were computerized based on the following grades: “no stress” 1″, “moderate stress” 2″, “stress” 3″, “very stress” 4″, “extreme stress” 5″. From these 6 domains, the researcher analyses which domain has the biggest impact on the level of stress by calculating the mean and standard deviation for each of the questions. For the second objective, the researcher calculated the total YES for each respondent and correlates it with each of the Section A questions (the domains with their respective items). The YES refers that the students have been physiologically affected whereas NO means the students are not affected physiologically by the items in each domains. Therefore, the researcher calculates the total yes for each respondent and correlates it with the domains. Thus, the stronger positive correlation which is closer to 1 indicates a strong relationship between their cause of stress and symptoms of psychological stress. For the third objective, the researcher intends to calculate the respondents’ locus of control whether is an external or internal locus of control. In the scale consisted of 29 items, there are two choices in every item a and b. Total point is between 0 and 23. Each of the questions in the questionnaire on Section 3 was given points accordingly and the total points are calculated. The range between 0-11 has an internal locus of control and range between 12-23 has an external locus of control. The maximum point is 23. Higher points show an increase in the external locus of control and lower points show that the individual has an internal locus of control. Then, the researcher tends to compare means between the locus of control with the six domains of the Section A questions to identify which domain is affected the most with an internal or external locus of control. The researcher computed the means for each domain by using the compute variable syntax in SPSS and compares it with the internal and external locus of control. The higher mean indicates higher stress by comparing with internal and external locus of control. If external has a higher mean then external locus of control tends to have a higher stress whereas if internal locus of control has a higher mean than they tend to have a higher stress compared to external. The results are discussed in the next chapter. For the statistical analysis of the study, IBM SPSS Statistics 23 (Statistical Package for the Social Sciences) was utilized. The level of significance in all analyses is considered 0.05. The researcher then analysed and interpreted the information provided by the informants in accordance with the objectives of the study.3.6 Limitation of surveyThe study was mainly focused on investigating their perceived stress and locus of control by students at the University of Malaya in Malaysia. In this study, the respondents might have biased their responses towards what perceptions or behaviours their social constructs promote. Besides, the survey might have some limitation where the sample size and lack of geographical coverage since the data collected only concentrated on one university. Therefore, the findings of this study cannot be generalized. Therefore, the researchers suggest replication of this study using larger samples from other parts of the world to validate the findings. Besides, the size of the sample is small where only 260 respondents. In opinion of the researcher, a bigger sample would probably enhance the reliability of the research. There is a possibility of some error to a limited extent. In a nutshell, this chapter begins with an introduction then describes the research design employed in this study. Population and sample study is also mentioned before discussing the research methodology used to conduct this research. In collecting data, library and online research were used. Quantitative method is chosen in which a set of questionnaires are used in the research procedure. The data that were collected are then analysed and discussed with the results are displayed in the following chapter.