Chapter 3: Research Methodology

Introduction
3.1 Research Design
3.2 Measurement of Variables/ Instrumentation
3.3 Data Collection
3.4 Sampling
3.5 Data Collection Procedures
3.6 Techniques of Data Analysis
3.7 Chapter Summary

3.0 Introduction

Chapter 3 will discuss about the procedures and methods carried out to collect data in this
research. In this chapter, it outlines all the details of research design applied, method used and
sampling design to collect data. Furthermore, it also highlights the research instrument,
constructs measurement as well as the techniques used to propose data analysis.

3.1 Research Design

Quantitative research is used to gain an understanding of reasons behind the hypothesis by

collecting data with the purpose of exploring new ideas for further quantitative research
(Creswell, 2013). While quantitative research is used to test the hypothesis from the
questionnaires formed, the data collected are based on predictive methods to deliver
statistical data and analyses the participant numerical relationships (Creswell, 2013). In this
research, numerical measurements and statistical analysis are being used, thus quantitative
research is being applied. Questionnaires will be used to accumulate quantitative data. This
technique allows us to study a large group of people, then make presumptions from the
samples being studied.

According to the research objectives, the aim of this research is to investigate the relationship
between independent variable (compensation, perceived supervisor support (PSS), perceived
organizational support (POS), and emotional exhaustion) and the dependent variable
(employees’ turnover intention). This research seeks to identify the cause-and-effect
relationship between the variables thus, descriptive research will be used to conduct this
research. Descriptive research refers to conclusive research which main objective is to
describe a situation and have a clear picture of the phenomena by collecting data and study
relevant samples. Descriptive research composed of creative exploration and serves to
organize the findings in order to fit them with explanations, and then test or validate that
explanation. In this case, descriptive research allows us to document and describe the
attributes of the study elements without influencing the study subject. Furthermore, the
interpretation of the data is simplified with the use of visual aids like charts and tables. Thus,
this research design is best suited for this research to establish that what factors affect
employees from the food and beverages service industry perceive turnover intention.

3.2 Data Collection Methods

Data collection was a systematic method of gathering data and measuring information from
various sources. It allows researchers to collect data for the research problem and test
hypothesis and evaluate the outcomes accurately. There are two sources for data collection
which are primary data and secondary data. In this research, researchers used both of the data
collection methods in conducting this research.

3.2.1 Primary Data

Primary data defined as an original research obtained by researcher on particular interested

variables through first-hand information for the specific purpose of the study (Sekaran &
Bougie, 2010). There are some methods of primary data collection such as self-administered
questionnaire, interview, focus groups and experimentation.

In this research, we will be using questionnaires to obtain results from target samples as our
primary data. The questionnaire was adopted and modified from several related research
study of the past researchers. Questionnaire is known to be cost effective research instrument
and convenient method to collect data because the questionnaires can be distributed to a
diverse population easily.

3.2.2 Secondary Data

Secondary data referred to gathering of data or information that existed and available in third
party sources (Sekaran et al., 2010). For examples, government publications, media, internet,
online articles and magazines, company records and others (Sekaran et al., 2010). For this
study, researchers mostly used secondary data for research’s literature review through
internet website and article. Most of the journal articles were searched via online database
system such as Google scholar, ScienceDirect and ResearchGate.

3.3 Sampling Design

According to Sekaran et al. (2010), sampling referred as the process of identifying and
determining the representative from the target population and right event in a fairly manner.
It is important to ensure the effectiveness and accuracy of the research being conducted.

3.3.1 Target Population

Target population is where researchers identified and targeted a group of people of interest
for purpose of research (Zikmund, W.G., Babin, B.J., Carr, J.C., & Griffin, M., 2013). The
researchers need a specific large group and take a sample to collect data in order to obtain
more reliable and accurate results on the relationship between IVs and DV. In this research,
our research objective is to investigate the factors that contributed to turnover intention
among employees in food and beverage service industry in Perak, Malaysia. Therefore, we
target on all the employees working in food and beverage service industry in Perak, Malaysia
such as cafés, beverage shop and food restaurants. The questionnaire is distributed to the
respondents in these industry sectors.

3.3.2 Sampling Frame and Sampling Location

The sampling frame referred to a set of elements where researcher can select a sample of the
target population. The sampling frame for this research is the employee in the food and
beverage service industry. Only the employee would be the respondent.

The sampling location means the location that the researcher chose to collect data. In this
research, the sampling location is in Perak, Malaysia. Based on GDP by State 2010-2016,
Department of Statistics, Malaysia, service sector in Perak ranked as number five out of the
states with 6.1% of share (Figure 3.1). Therefore, we take Perak as our sampling location.
 Figure 3.1: GDP by State 2010-2016.

Source: Department of Statistics, Malaysia

3.3.3 Sampling Element

Sampling element is referred to the respondent in this research. For this research, we target
the employees in food and beverage service industry such as waiter/waitress, manager,
operators and other positions working in cafes or beverage shop and food restaurant. We
select 300 employees to fill up our questionnaire.

3.3.4 Sampling Technique

Sampling technique consists two categories which are probability sampling and non-
probability sampling. This research used non-probability sampling. There are three types of
sampling under non-probability sampling which are the convenience, purposive and quota
sampling. In this research, we chose convenience sampling to sample our population because
it enables us to obtain information easily together with cost and time effective. This sampling
technique was suitable for research since some of the participated café or restaurant gave no
permission to conduct questionnaire distribution in systematic count for every participated
employee.

3.3.5 Sampling Size

Sampling size is the number of respondent collected in the research. The adequate sample
size must be more than 30 and between 500 (Sekaran & Bougie, 2013). The sample size
cannot exceed 500 as it may cause a problem which is Type II errors as possible (Sekaran et
al., 2013). Therefore, we chose 300 respondents as our sample size and we will distribute 300
questionnaires to the targeted subject. Besides that, we also chose 30 set of questionnaire as a
pre-test to ensure the quality, reliability and validity of the questionnaire.

3.4 Research Instrument

Research instrument is defined as a tool that facilitates data collection, measurement and data
analysis in completing the research. In this research, questionnaire is used to gather the
responses from the potential respondents on understanding the factors that contribute to
employee turnover intention among employees in food and beverage service industry in
Perak, Malaysia. The respondents are required to follow the instructions stated in the
questionnaire in completing the survey.

3.4.1 Questionnaire Design

In the study, the questionnaire is distributed to the target population via online using Google
Form as well as walk-in method and close-ended questions are asking in the questionnaire.
English is used as the main language in presenting the questions while Malay version
questions are attached together to help those respondents who are having difficulties in
understand English to answer the questionnaire.

The questionnaire consists of 30 questions and it is divided into three sections which are
Section A, B and C. Section A consists of 5 questions aimed to collect the basic demographic
information of the respondents in the basis of gender, age, level of academy, working
experiences and job position. Section B and C are designed using five pointing Likert scale
which indicates strongly disagree, disagree, neutral, agree and strongly agree that reflect
opinion of the respondents on the factors that contribute to employee turnover intention
among employees in food and beverage service industry in Perak, Malaysia. There are 5
questions about DV of the research, turnover intention in Section B and 5 questions per each
IVs which are perceived supervisor support (PSS), perceived organizational support (POS),
compensation and emotional exhaustion in Section C.

3.4.2 Pilot Test

Pilot test is a pre-test of the research that allows the researchers to collect data from a small-
sized and representative sample of respondents before conducting the actual survey (Wright,
2018). It allows the researchers to identify the potential errors in the questionnaire and ensure
the accuracy and reliability of the questionnaire. 30 respondents from the target population
are chose to conduct the pilot test and the responses are tested for reliability and validity of
the questionnaire through Statistical Package for Social Science (SPSS) version 25.0. Table
3.1 shows the result of the pilot test.

Table 3.1: Pilot Test's Cronbach's Outcomes

Number of Cronbach’s Level of

Variables Dimension
Items Alpha Reliability

Dependent Turnover
5 0.956 Strong
Variable Intention

Perceived
Supervisor 5 0.932 Strong
Support

Perceived
Independent Organizational 5 0.876 Strong
Variable Support

Compensation 5 0.871 Strong

Emotional
5 0.853 Strong
Exhaustion

Average of Cronbach’s Alpha 0.898 -

Source: Developed for the research.

3.5 Construct Measurement

According to Sekaran and Bougie (2016), scale is defined as a tool, mechanism or instrument
by which one are differential from one another based on the variables of interest in a
meaningful way. It helps the research to determine and develop the accurate or measureable
value.

3.5.1 Scale Measurement

There are four levels of scale measurement which are nominal scale, ordinal scale, ratio scale
and interval scale (Sekaran & Bougie, 2016).

3.5.1.1 Nominal Scale

Nominal scale is the scale measurement which classifies information into categories without
any numerical representation (Willits, Theodori, & Luloff, 2016). For example, gender, the
general information question in this research is using nominal scale. The answer for the
question is classified as either male or female. The figure shown below is the example
questionnaire using nominal scale (Figure 3.2).

Figure 3.2: Example of nominal scale.

Source: Developed for the research.

3.5.1.2 Ordinal Scale

Ordinal scale refers to the measurement by ranking or order of the representative categories
or items without information concerning the distance between them (Willits, Theodori, &
Luloff, 2016). Age group of the employees in the research is one of the example questions
using ordinal scale measurement (Figure 3.3).
 Figure 3.3: Example of Ordinal Scale.

Source: Developed for the research.

3.5.1.3 Interval Scale

Interval scale is the combination of both nominal scale and ordinal scale (Sekaran & Bougie,
2013). It is a numerical scale that provides the actual difference between order and value by
calculating mean, median, mode and standard deviation of the data. According to Sekaran
and Bougie (2013), Likert scale is considered as part of the interval scale. In the research, the
Likert scale used in the questionnaires is the example of interval scale (Figure 3.4). Strongly
Disagree is indicated as pointer 1 and Strongly Agree is indicated as pointer 5 in Likert scale
while Disagree, Neutral and Agree are indicated as pointer 2, 3 and 4 respectively.
 Figure 3.4: Example of Interval Scale.

Source: Developed for the research.

Respondents are asked to provide their opinions for each statement by ticking [√] in the
appropriate box labelled with pointer. The following figure (figure 3.5) represents the sample
question model.

Figure 3.5: Example question model.

Source: Developed for the research.

3.5.2 Operational Definition of Constructs

3.5.2.1 Turnover Intention

Turnover intention means that the employees intend to leave the organization.

Table 2.2: The Assessment for Turnover Intention

NO Statement SD D N A SA
6 I often think of leaving my current workplace. 1 2 3 4 5
Saya sering berfikir untuk meninggalkan tempat
kerja saya sekarang.
7 I am planning to work at somewhere else. 1 2 3 4 5
Saya sedang merancang untuk bekerja di tempat
lain.
8 I am bored with my current position and planning 1 2 3 4 5
to look for a new one.
Saya bosan dengan jawatan semasa dan
marancang untuk cari yang baru.
9 There is a big possibility that I will look for a new 1 2 3 4 5
job next year.
Kemungkinan besar saya akan cari pekerjaan
baru pada tahun depan.
I0 I will leave my current workplace if I got a better 1 2 3 4 5
job offer.
Saya akan meninggalkan jawatan semasa
sekiranya mendapat tawaran yang lebih baik.
Source: Developed for the research.
3.5.2.2 Perceived Supervisor Support (PSS)

Perceived Supervisor Support (PSS) defines the general perception of employees about the
rate supervisor values them and cares for their welfares.
Table 3.3: The Assessment for Perceived Supervisor Support (PSS)

NO Statement SD D N A SA
11 My supervisor cares about my well-being. 1 2 3 4 5
Penyelia saya prihatin dengan kesejahteraan /
kebajikan saya.
12 My supervisor cares about my opinions. 1 2 3 4 5
Penyelia saya prihatin dengan pendapat saya.
 13 My supervisor is available when help is needed. 1 2 3 4 5
Penyelia saya ada semasa pertolongan
diperlukan.
14 My supervisor involves my participation in 1 2 3 4 5
decision making.
Penyelia saya melibatkan penyertaan saya
apabila membuat keputusan.
15 My supervisor evaluates my work and the 1 2 3 4 5
feedback is useful.
Penyelia saya menilai kerja saya dan maklum
balas diberi adalah berguna.
Source: Developed for the research.

3.5.2.3 Perceived Organizational Support (POS)

Perceived Organizational Support (POS) is the degree to which employees believe that their
organization values their contribution and cares about their well-being and fulfills their socio-
emotional needs.
Table 3.4: The Assessment for Perceived Organizational Support (POS)

NO Statement SD D N A SA
16 My general satisfaction at work is taken cares by 1 2 3 4 5
the company.
Kepuasan umum saya di tempat kerja dijaga oleh
syarikat.
17 The organization greatly appreciates my 1 2 3 4 5
contribution to its well-being.
Sumbangan saya terhadap kesejahteraan
organisasi dihargai oleh syarikat.
18 My goals and values are considered by the 1 2 3 4 5
company.
Matlamat dan nilai saya dipertimbangkan oleh
syarikat.
19 I was offered with the best job for which I am 1 2 3 4 5
qualified.
 Saya ditawarkan dengan pekerjaan terbaik yang
saya layak.
20 The job given for me was interesting. 1 2 3 4 5
Pekerjaan yang diberikan untuk saya menarik.
Source: Developed for the research.

3.5.2.4 Compensation

Compensation is the total cash or non-cash payments the employees get in exchange for the
work they do for the business.
Table 3.5: The Assessment for Compensation

NO Statement SD D N A SA
21 I am satisfied with salaries, wages or incentive 1 2 3 4 5
given by the company.
Saya berpuas hati dengan gaji, upah atau insentif
yang diberikan oleh syarikat.
22 Fair compensation or allowance is given based on 1 2 3 4 5
performance.
Ganjaran atau elaun yang adil diberikan
berdasarkan prestasi.
23 I am being appreciated for my contribution. 1 2 3 4 5
Saya mendapat penghargaan yang setimpal atas
sumbangan saya.
24 I am satisfied with the available facilities and 1 2 3 4 5
infrastructure.
Saya berpuas hati dengan kemudahan dan
prasarana yang disediakan.
25 I able to see opportunities for my career and 1 2 3 4 5
work.
Saya dapat melihat peluang untuk kerjaya dan
pekerjaan saya.
Source: Developed for the research.

3.5.2.5 Emotional Exhaustion

Emotional Exhaustion is the state of feeling emotionally worn-out and drained as a result of
accumulated stress from personal or work lives, or the combination of both.
Table 3.6: The Assessment for Emotional Exhaustion

NO Statement SD D N A SA
26 I feel stressed working on my job. 1 2 3 4 5
Saya berasa sangat tertekan ketika melaksanakan
tugasan.
27 I feel tired to face my job every day. 1 2 3 4 5
Saya berasa letih untuk menghadapi pekerjaan
saya setiap hari.
28 I feel strained working with people all day. 1 2 3 4 5
Saya tertekan semasa bekerja dengan orang
sepanjang hari.
29 I feel emotionally exhausted from my job. 1 2 3 4 5
Saya berasa letih dari segi emosi terhadap
pekerjaan saya.
30 I feel mentally and physically burned out end of 1 2 3 4 5
the day.
Saya berada dalam keadaan burnout dari segi
mental dan fizikal selepas habis kerja.
Source: Developed for the research.
3.5.2.6 Origin and Measure of the Construct

Table 3.7: Origin of Construct

Construct Statement Sources no of items

Turnover 1. I often think of leaving Parashakti, R. D., Nashar, 5
Intention (DV) my current workplace. M., & Usliawati, D.
2. I am planning to work (2017). The Effect of
at somewhere else. Compensation and
3. I am bored with my Organization
current position and Commitment towards
planning to look for a Turnover Intention. Case
new one. Study in ILC
4. There is a big (International Language
possibility that I will Center) Jakarta. In 2017
 look for a new job International Conference
next year. on Organizational
5. I will leave my current Innovation (ICOI 2017).
workplace if I got a Atlantis Press.
better job offer.
Perceived 1. My supervisor cares Ghazali, H., Nashuki, N. 5
Supervisor about my well-being. and Othman, M. (2018).
Support (PSS) 2. My supervisor cares The Influence of
about my opinions. Perceived Organizational
3. My supervisor is Support (POS), Perceived
available when help is Supervisory Support
needed. (PSS) and Organizational
4. My supervisor Commitment (OC)
involves my towards Intention to
participation in Leave or Intention to
decision making. Stay: A case of Casual
5. My supervisor Dining Restaurants in
evaluates my work Klang Valley,
and the feedback is Malaysia. International
useful. Journal of Academic
Research in Business and
Social Sciences, [online]
8(9).
Perceived 1. My general Gök, Ö. A., Akgündüz,
Organizational satisfaction at Y., & Alkan, C.
Support (POS) work is taken cares (2017). The effects of job
by the company. stress and perceived
2. The organization organizational support on
greatly appreciates turnover intentions of
my contribution to hotel employees. Journal
its well-being. of Tourismology, 3(2),
3. My goals and 23-32.
values are
considered by the
 company.
4. I was offered with
the best job for
which I am
qualified.
5. The job given for
me was
interesting.
Compensation 1. I am satisfied with Parashakti, R. D., Nashar,
salaries, wages or M., & Usliawati, D.
incentive given by the (2017). The Effect of
company. Compensation and
2. Fair compensation or Organization
allowance is given Commitment towards
based on performance. Turnover Intention. Case
3. I am being appreciated Study in ILC
for my contribution. (International Language
4. I am satisfied with the Center) Jakarta. In 2017
available facilities and International Conference
infrastructure. on Organizational
5. I able to see Innovation (ICOI 2017).
opportunities for my Atlantis Press.
career and work.
Emotional 1. I feel stressed working Maslach Burnout
Exhaustion on my job. inventory – General
2. I feel tired to face my Survey (MBI-GS)
job every day. (Schaufeli et al., 1996)
3. I feel strained working
with people all day.
4. I feel emotionally
exhausted from my
job.
5. I feel mentally and
physically burned out
 end of the day.
Source: Developed for the research.
3.6 Data Processing

Data processing is performed after completing the collection of questionnaires. It ensures the
collected data is able to create high quality research project by providing the researchers with
more accurate data and creating lesser error. It involves data checking, data editing, data
coding, data transcribing and data cleaning.

3.6.1 Data Checking

Data checking is used to find out and eliminate the potential errors in the questionnaires and
results obtained from the respondents. The errors include inappropriate wording in the
questions, grammar mistake, confusing layout of the questions and uncompleted
questionnaires. The process is aimed to maintain the reliability of the research.

3.6.2 Data Editing

After checking the data, the data is edited to ensure the consistency and accuracy of the
questionnaires. The data collected from the respondents might be illegible, incomplete or
ambiguous. Hence, the collected data is monitored and adjusted to improve the quality of the
overall statistical research study.

3.6.3 Data Coding

Data coding is the step where numbers or symbols are assigned to the responses collected
from respondents to be entered into Statistical Package of Social Science (SPSS) software
version 25.0 to ease the interpretation of data. Data entry became more effective and efficient
using data coding as compared to lengthy alphabetical descriptions. The questionnaires of
this research consist of three sections and are divided into two parts in data coding process.
Section A which is the general information of the respondents is categorised as part A is
shown in Table 3.8 and the other 2 sections which aimed to study for the dependent and
independent variables of the research are categorised as part B is shown in Table3.9.
 Table 3.8: Data coding for Section A

Question Label Coding

1 Gender 1 = Male
2 = Female
2 Age 1 = 24 years and Below
2 = 25 years – 34 years
3 = 35 years – 44 years
4 = 45 years – 54 years
5 = 55 years and Above
3 Level of Academy 1 = SPM / STPM
2 = Diploma
3 = Bachelor’s Degree
4 = Master’s Degree
5 = Doctorate
6 = Others
4 Working Experiences 1 = 2 years and Below
2 = 3 years – 5 years
3 = 6 years and Above
5 Job Position 1 = Manager
2 = Cook / Barista
3 = Operator (Dishwasher / Server / Cashier)
Source: Developed for the research.

Table 3.9: Data coding for Section B

Question Label Coding

5 Questions Dependent Variable 1 = Strongly Disagree

(DV): Turnover 2 = Disagree
Intention 3 = Neutral
4 = Agree
5 = Strongly Agree

20 Questions Independent Variables 1 = Strongly Disagree

 (IV): 2 = Disagree
3 = Neutral
(5 Questions for  Perceived
4 = Agree
each IV) Supervisor
5 = Strongly Agree
Support (PSS)

 Perceived
Organizational
Support (POS)

 Compensation

 Emotional
Exhaustion

Source: Developed for the research.

3.6.4 Data Transcribing

Data transcribing is the step where coded data is transformed directly into the computer using
Statistical Package of Social Science (SPSS) software version 25.0. The data will then being
interpreted and analysed.

3.6.5 Data Cleaning

The last step in data processing is data cleaning. It is aimed to verify and identify the data
which is out-of-range, logically inconsistent or have extensive values as to increase the
reliability and consistency of the data.

3.7 Data Analysis

Data Analysis is the process which all the data are collected through questionnaires and will
be interpreted into useful information (Zikmund, W.G., Babin, B.J., Carr, J.C., & Griffin, M.,
2013). The data will be described and illustrated, condensed and recapped and finally
providing meaning to the collected data by applying statistical or logical techniques
systematically. The accuracy and reliability of the data is important to ensure the integrity of
the data and avoid misled for the readers.

In this research, Statistical Package for Social Science (SPSS) 25.0 is used to examine the
data collected in order to test the reliability, significant of data, descriptive analysis and other
accuracy of measurement of data. Hypotheses findings of this research can be evaluated and
supported using SPSS as well as investigating the relationship between the independent
variables and dependent variables using methods such as Pearson Correlation analysis and
Multiple Regression analysis.

3.7.1 Descriptive Analysis

According to Sekaran (2016), descriptive analysis is the measuring of frequencies, central of

tendency (mean, median and mode) and dispersion of variables (standard deviation and
variance). To get the feel of data, summarization of the output of data can be conveyed in
visualized ways such as: chart, graph, table, histogram and etc. This enable the ease to
compare and describe the data collected. The Statistical Package of Social Sciences (SPSS)
version 25.0 had been used to analyse and interprets the data from gathered questionnaires.

3.7.2 Scale Measurement (Cronbach’s Alpha Reliability Test)

Reliability analysis in the scale measurement is used to test if the raw data can generate
reliable and consistent results. According to Sekaran (2016), consistency represented as
proficiency or “goodness” of the measure. Reliability refers to which the measures are free
from errors and have a stable result. Cronbach’s Alpha reliability test is widely used to
measure the internal consistency reliability between different items in questionnaire. It
measures how positively correlated among each items with the range of 0 (no consistency) to
1 (perfect consistency) where the closer the number to 1, the higher the internal consistency
reliability. Thus, Cronbach’s Alpha was computed to assess the internal consistency
reliability of the four independent variables (compensation, perceived supervisor support
(PSS), perceived organizational support (POS), and emotional exhaustion).

Table 3.10: Cronbach’s Alpha Coefficient sizes

Cronbach’s Alpha Value Reliability

More than 0.8 Strong
Above 0.6 Moderate
Less than 0.6 Weak
Source: Developed for the research.
3.7.3 Inferential Analysis
Inferential analysis makes assumptions or predictions on a population through observation of
the sample analysis. The result of sample analysis can be generalized to the whole population
and act as the representatives or signify of the particular sample population. All of the
hypothesis in this research will be tested using inferential analysis, Pearson Correlation
Coefficient and Multiple Regression Analysis which are included under inferential analysis

3.7.3.1 Pearson Correlation Coefficient

According to Sekaran (2016), Pearson correlation indicates relationships as well as casual

effect between the variables. Pearson correlation indicates how strong or weak the
relationship between the independent variables and dependent variable. The closer it is to 1,
the stronger the relationship (positive relationship). Based on Sekaran (2016), it is possible to
have negative figure -1 and inverse relationship (negative relationship). The general
acceptable level of significant correlation level in social science is where P = 0.05 with the
range from -1.00 to +1.00. This analysis is focused on the strength and direction of
relationship among the four independent variables which are compensation, perceived
supervisor support (PSS), perceived organizational support (POS), and emotional exhaustion.

Table 3.11: Pearson Correlation Coefficient range

Coefficient Range Strength of Relationship

0.01 – 0.20 Very weak
0.21 – 0.40 Weak
0.41 – 0.70 Moderate
0.71 – 0.90 Strong
0.91 – 1.00 Very strong
Source: Developed for the research.

3.7.3.2 Multiple Regression Analysis

According to Sekaran (2016), multiple regression analysis is being used when there is more
than 1 independent variable which will have impact towards the dependent variable. Multiple
regression analysis is to test the relationship between dependent variable and independent
variables by regress the coefficient. It will represent the relative importance in predicting the
dependent variable.
All the independent variables are inserted into the regression to be regressed by dependent
variable for explaining variance. Every single coefficient represents the relationship of
independent variable with dependent variable. It also explains how increasing or decreasing
of 1 unit in the coefficient will affect the dependent variable as well (Sekaran, 2016).

The Multiple Regression Model equation is as follow:

Y = a + βX1 + βX2 + βX3 + βX4

Where,

Table 3.12: Multiple Regression Model items

Equation Symbol Variable Label

Y Dependent Turnover Intention
Perceived supervisor support (PSS)
X1 Independent 1
4
Perceived organizational support
X2 Independent 2
(POS) 3
X3 Independent 3 Compensation 1
X4 Independent 4 Emotional Exhaustion 2
Source: Developed for the research.
a = the intercept of the regression line or constant point where the straight line intersects the
y- axis (when x=0)

β = the slope of the regression line or regression coefficient for X (the change in y for every
1unit change in x, subject to other variables remain constant)

3.8 Conclusion

In a nutshell, this chapter explained how the research is carried out in term of sampling
design, data collection methods, research instruments, data processing and data analysis
process. The results obtained from this chapter will be further discussed in Chapter 4. All the
hypothesis testing had been performed.

4.2.1 Interactional Justice

Table 4.17: Reliability Statistics- Interactional justice

Cronbach's
Alpha Based
on
Cronbach's Standardized
Alpha Items N of Items

.771 .800 6