CHAPTER 3

RESEARCH METHODOLOGY

3.0 Introduction
In the earlier chapters, the researcher has looked upon the viewpoints of
various researchers’ related to our study titled “A Study of Employee Motivation
Practices in Fast Food Sector”. In this chapter, the researcher will detail the methods
used in carrying out our study. We will discuss about the research design for our
study, different methods of data collection such as primary data and secondary data.
Later on, the researcher will determine the sample design, research instruments and
analytical tools being used in our data collection and analysis. Further the data
processing and data analysis will be described. In short, this chapter will illustrate
the way data are being collected as well as the data analysis methods in interpreting
our survey results.

3.1 Research Design

Research design provides the fundamental direction to the research project
for conducting the study.

According to Zikmund (2000), the major underlying principle of research

design includes exploratory, descriptive and causal researches, which are commonly
known as explanatory research. Exploratory research conducts into a research
predicament or issues which few or no previous studies we can refer for information.

The next research method is descriptive research. It is conducted to explain

the obtainable phenomena. This approach is to recognize and obtain information on
the distinctiveness of a particular problem or issue. Further descriptive research
examines a problem rather than exploratory research, as it is undertaken to describe
the characteristics of the related issues, objects, or people.

Another type of research is causal research. It is the type of study that

pinpoints the cause and effect relationship between the variables. It also attempts to
create a connection between independent variable and dependent variable.
Chapter 3- Research Methodology 36

As an outcome, the researcher has selected descriptive and causal research

for our study as descriptive research can be used to express the respondents’ data,
such as type of employment, work experience, gender, age, race, marital status, and
level of education, whereas causal research can be used to make a diagnosis of the
cause and effect relationship between independent and dependent variable.

Furthermore, there are two basic approaches that can be used to investigate
the data, which are quantitative and qualitative approach. Qualitative research is
defined as broader term for analytical methodologies described as naturalistic,
anthropological or observation based research. It is differed from quantitative
research which attempts to gather data by objective methods for the information
about relations, comparisons, and predictions and attempts to eliminate the
researcher from the research (Smith, 1983). According to Denscombe (1998), the
difference between qualitative and quantitative approach is that the data focuses on
either words or numbers. Qualitative mainly focus on observation of data while
qualitative focus on measurement of numbers and data.

The main rationale of this research is to find out the motivational factors
affecting the employees’ job in fast food industry. We use quantitative data
collection methods to investigate and collect information. As per Lodico, Spaulding
and Voegtle (2010), quantitative research enables the researcher to analyse the most
influential factor that affects dependent variable.

Lastly, the unit of analysis in this study is focusing on the employees in fast
food industry, such as Domino’s, KFC, McDonalds’, Burger King and Pizza Hut in
New Delhi, India. Total up of 200 sets of questionnaires to be distributed or e-
mailed to the selected targeted group fast food employees. The researcher design the
questionnaire with close-ended questions on a seven point scale as it encourages
respondents’ readiness to complete the survey. Moreover, this type of questionnaire
consumes less time, cost and energy.
Chapter 3- Research Methodology 37

3.2 Data Collection Methods

There are two types of data collection method being used in the current
study, which are primary data and secondary data.

3.2.1 Primary Data

Primary data is defined as data that has not been previously published as the
data is derived from a new or original research study. According to the researchers
(Hair et al., 2007), a research cannot be accomplished with only depending on
secondary data. Researchers are required to collect and gather primary data to obtain
a more accurate result. Primary data collection method can be divided into two
different methods, which are qualitative and quantitative research (Hair et al., 2007).
In this research, the researcher has used the quantitative technique which is
questionnaire survey form. This method is being selected so as to enable the survey
to be completed in a convenient and cost efficient way. Moreover, the data obtained
is original as it was obtained directly from the target sample.

Questionnaire has been chosen in collecting primary data. The questionnaires

were distributed to respondents through face to face as well as e-mail. Collection
data electronically is an effective way to collect data from a large sample of working
people within a short period. All the responses from the respondents are kept in
private and confidential.

3.2.2 Secondary Data

Secondary sources can be defined as a description, interpretation, analysis,
evaluation, explanation, or comment on a primary source or event. According to
Hair et al. (2007), secondary data is the data which is already collected by other
researchers and is available for others to be used. We have collected secondary data
from various journals, newspapers, websites and books etc. The main reason to use
journals is to collect the relevant information provided by previous researchers. This
method is considered as a time-efficient way for us to discover new information
relevant to our study.
Chapter 3- Research Methodology 38

3.3 Sampling Design

Sampling is a course of action of selecting a satisfactory number of elements
from a population. It increases the probability of getting the results from the sample
to be more attributable to the population.

3.3.1 Target Population

According to Sekaran and Bougie (2010), population is the total number of
people, events or things that researchers wish to investigate on. As stated in the
study in 2016 by FICCI’s KPMG on “India’s food service industry: Growth
receipe”, “The Gross Value Added (GVA) by hotels and restaurants has witnessed
an annual increase in both absolute and relative terms. It grew from INR1,084.2
billion in FY14 to INR1,211.7 billion in FY15, growing its contribution in country’s
overall GVA from 1 per cent in FY14 to 1.1 per cent in FY15. The restaurant
industry is expected to contribute about 2.1 per cent to the total GDP of India by
2021. The F&B sector has evolved over the past decade, giving rise to exciting new
concepts in food and beverage offerings and new and innovative service elements.”

Fast Food Chains in India have been on the rise in the past few years. Some
of the leading fast food chains in India are Subway, McDonald's, KFC and
Domino's. The researchers studied that a number of fast food restaurants are doing
very well in the industry in India. In terms of number of outlet, as at 2016, the top
five leading food service chain brands in India are Domino’s Pizza (1106 outlets),
Pizza Hut (360 outlets), KFC (350 outlets), McDonald’s (296 outlets) and Burger
King (103 outlets).

As a result, all the workers that involved in the fast food industry will make
up the population for this study.

3.3.2 Sampling Frame and Sampling Location

Sekaran and Bougie (2010) stated that sampling frame is the sample that is
drawn from a list of population elements that usually might be different from the
target population in actual practices.
Chapter 3- Research Methodology 39

As per the KPMG report (2016) Fast Food Chains in India have been on the
rise in the past few years. Some of the leading fast food chains in India are Subway,
McDonald's, KFC and Domino's. Fast Food Chains in India offer items that include
vegetarian and non vegetarian dishes like pizzas, burgers, cheese dip, chicken wings,
fried chicken, and French fries. Subway is one of the leading fast food chains in
India and it opened its first restaurant in the country at New Delhi in December,
2001. It has grown over the years to have more than hundred restaurants that are
spread all over the country.”

Source: NRAI/Technopak, http://www.thehindubusinessline.com/ news/variety/indias-restaurant-

revolution/article4720776.ece

Figure 3.1: Food Service Market Size in India

The sampling frame for the research is the employees who work or worked
in the fast food service sector. The researcher narrow down the amount of
employees by focusing only on those who have work experience in Domino’s,
Burger King, McDonald’s restaurant, KFC restaurant and Pizza Hut restaurant. For
the sampling location, the researcher targeted on the fast food restaurants located in
New Delhi. Overall, the results obtained from these locations are believed to be
generalized to the population.
Chapter 3- Research Methodology 40

3.3.3 Sampling Elements

The respondents for the study are the employees who work or worked in
McDonald’s, Domino’s, Burger King, KFC and Pizza Hut restaurants. The
researcher targeted the employees at all the hierarchical levels from these fast food
restaurants, which include part-timer to permanent staff as well as management level
workers to a general worker.

3.3.4 Sampling Technique

There are two types of sampling method which are probability sampling and
non-probability sampling. In probability sampling, each of the elements in the target
population has an equal probability of being chosen as the sample for the survey
conducted. Probability sampling is convenient and simple in theory, and the results
obtained from this method are more generalizable towards the target population. For
non-probability sampling, each of the elements in the sampling frame does not have
an equal chance to be chosen as the sample. Admittedly this method is simpler and
convenient to operate however the results obtained cannot be confidently
generalised to the population. For this study, probability sampling was selected as
this method could assure the result obtained is representative toward the population.

3.3.5 Sampling Size

According to the data getting from scholars’ researches, the population for
our study is large which accounts for around 22,000. Based on the study of Krejcie
and Morgan (as cited in Sekaran and Bougie, 2010), the appropriate sample size for
the population of 22,000 should be 227, hence we round up the figure to 200.
Further, we decide to distribute 230 questionnaires to respondents to avoid the
happening of some incident which may reduce the sample size and therefore affect
the final results that is unreturned questionnaire from respondents.

Primarily, the researcher distributed 35 sets of survey questionnaire for the

purpose of pilot test. After the researcher collected back all the 35 sets of
questionnaire and ensured the reliability of all the constructs, the actual survey was
started by distributing 200 sets of survey questionnaire to the employees who work
or worked in McDonald’s, Domino’s, Burger King, KFC and Pizza Hut restaurants.
Chapter 3- Research Methodology 41

3.4 Research Instrument

According to Zikmund, Babin, Carr, and Griffin (2010), questionnaire is a
system for collecting data which requires the respondents to answer the same set of
questions in a predetermined order. As compare to interview and observation,
questionnaire is believed to be the most suitable method to collect considerable data.
Questionnaire has been chosen in this research as it can help the researcher to
pinpoint the most prominent factors that affect employees’ motivation practices.

3.4.1 Questionnaire Design

The research questionnaire consists of close ended questions with fixed
alternative questions. Close ended questions are the questions that provide
respondents with multiple or limited choices to answer. Fixed-alternative questions
limit the respondent to possible replies only and thus avoid bias introduced by
articulate individuals’ longer answers. It is uniform thus requires less interviewer’s
skill, takes less time to answer, easier for the respondents to answer and provide
comparability of answers. Basically, there are two different types of fixed-
alternative questions as being used in this research, one is simple-dichotomy
question and another one is determinant-choice question. Simple-dichotomy
question which requires respondents to choose one out of two alternatives, while
determinant-choice question which requires respondents to choose one of several
alternatives. Further we have also used 7 point scale in the questionnaire.

Example of Simple-Dichotomy Question

Are you employed or unemployed?

Employed Unemployed

Example of Determinant-Choice Question

What is your income?

Upto 10K 10k-15K 15K-25K 25K-35K Others (Please specify):
___________________________
Chapter 3- Research Methodology 42

3.4.2 Pilot Test

Before we carry out the actual survey, a pilot test has been run to test the
design of our questionnaire. Pilot testing is a small-scale trial, where a few
examinees take the test and comment on the mechanics of the test. They point out
any problems with the test instructions, instances where items are not clear and
formatting and other typographical errors and/or issues. The main purposes of this
pilot test are to test the respondents’ understanding of question meaning as well as
the question precision. It is also used to check the reliability and internal
consistency.

In pilot test, 20 respondents were being chosen from the employees who are
working in fast food industry in New Delhi. There were 15 respondents who
completed the questionnaire through online and the other 5 were given the survey
questionnaires directly and they filled it on the spot. Data collected from the
respondents is essential for identifying possible problem and error occur in the
questions, hence rectification can be made for the assurance of more significant and
accurate data. The pilot test had been completed in 1 week before we proceed to the
distribution of questionnaire to the real sample size. The result of reliability test for
the pilot test is shown in Table 3.1 and the interpretations for the result will be
discussed in the following chapter Chapter 4 “Results and Discussions”.

Table 3.1: Pilot Test’s Cronbach’s Alpha Result

Construct Cronbach’s Alpha Number of Items
Motivation 0.811 5
Personal Factors 0.718 5
Training 0.756 6
Job Characteristics 0.712 5
Job Involvement 0.838 6
All Variables
(Motivation, Personal 0.883 27
Factors, Training, Job
Characteristics, and Job
Involvement)
Source: Developed for the research
Chapter 3- Research Methodology 43

3.5 Constructs Measurement (Scale and Operational Definitions)

The questionnaire consists of several questions. It contained questions
regarding the factors affecting the employees’ personal particulars, job involvement,
training needs, job characteristics, motivation etc. The questions A are used to
measure the independent variables and dependent variable (Motivation, Personal
Factors, Training, and Job Characteristics, and Jon Involvement). As a result, all the
questions are designed with interval scale. The interval scale is a standard rating
scale which interprets differences in the distance along the scale (example:
extremely satisfied to extremely dissatisfy). This kind of scale is best suit for
opinion or attitude measurement. The scaling technique of interval measurement
scale being used in our questionnaire is 7-Point Likert Scale, in which there are
inconclusive results on the use of a middle or neutral point.

According to Alreck, Selttle, and Miller, Likert scale is more consistent and
suitable. This kind of Likert-scale is selected based on the recommendation of
Kronsnick and Fabrigar. They clarified that the range of scale between 5 and 7 is set
up to be more consistent than five-point Likert scale and the further argued that 7-
point scale appears to be most favorable when measuring a bipolar. This is due to
the reason that the scale with more points enables the respondents to express their
viewpoint more specifically and comfortably. In addition, seven-point Likert scale
allows the researcher to make more subtle division among the various respondents’
attitudes regarding to a particular item.
Disagree
Disagree

Disagree
Strongly

Strongly
Neutral
Slightly

Slightly
Agree
Agree

Agree
Sr.
Questions
No.

1. I am satisfied with my benefit package. 1 2 3 4 5 6 7

Participating in training help me increase my

2.
salary.

3.5.1 Standard Measurements of Independent Variables

There are five independent variables (Motivation, Personal Factors, Training,
and Job Characteristics) will be measured by adapting the previous researchers’
standard measurements.
Chapter 3- Research Methodology 44

3.5.1.1 Motivation
There eight questions in order to test the impact of motivation towards
employees’ job involvement. The questions under motivation part were adapted
from “Employee Motivation Questionnaire” which was developed by Fourie (1989).
We chose the questions that are considerably related to the employees’ job
involvement. Also, we extend the Likert scale to seven-point Likert scale from five
point scale in the original questionnaire.

3.5.1.2 Training
There are five questions modified to test the impact of training towards
employees’ job involvement. All the questions under training part were developed
by Noe and Wilk (1993) and designed in five-point Likert scale originally. Again,
we converted it to seven-point Likert scale.

3.5.1.3 Job Characteristics

There are five questions adapted to test the impact of job characteristics
towards employees’ job involvement. All the questions under job characteristics
were developed by Hackman and Oldham and they were designed in five-point
Likert scale. We extend it seven-point Likert scale.

There are nine questions adopted to test the employees’ job involvement. All
the questions under job involvement were developed by Kanungo (1982) and they
were designed in five-point Likert scale. Similarly, we extend the five-point Likert
scale to seven-point Likert scale.

3.6 Data Processing

After collecting all the questionnaires from the respondents, data processing
step is then taken before any analysis being implemented. The data processing
consists of checking, editing, coding, classifying, tabulating and transcribing.
Meanwhile, all the unusual responses are identified.

Before checking all the collected questionnaires, the researcher counted and
numbered every questionnaire to assure the required amount of questionnaires is
being returned by the respondents. After that, all the data was checked that was
Chapter 3- Research Methodology 45

collected to make sure that all the questions in the questionnaire were being filled up
entirely by the respondents without any omission. Any incomplete questionnaire
found would then be taken out. Next, was coding, data entry, editing, and data
conversion for all the remaining survey questionnaires.

3.6.1 Coding
This step is taken in order to key in the data into the Statistical Package for
Social Sciences (SPSS) system. The researcher assigned the code to each
participant’s answer. For instance, in personal Particulars, under type of
employment, the researcher assigned “0” to part-time worker, “1” to full-time
worker.

Apart from that, the researcher attributed “0” for male and “1” for female.
However, for the rest questions, we started the code from “1” instead of “0” for all
the responses. For example, the researcher attributed from “1” to “8” for all the
responses of age.

Furthermore all the responses for the independent variables and dependent
variable were coded with the scale ranging from “1” to “7”. “1” symbolises strongly
disagree and “7” symbolises strongly agree. At times, all the missing data were
coded as “99.”

3.6.2 Data Entry and Editing

Once all the questions’ responses had finally been coded, the researcher
began to enter all the data into the SPSS database. This process is also known as
transcribing (Malhotra, 1993). Before running the reliability test, the researcher
carried out the editing tasks for all the responses of the questionnaires. The
researcher tried to spot and correct the problems, such as illogical, conflicting, or
illegal responses. Inconsistent responses happened when the respondents’ responses
that are incoherent with other information provided. Also, it is possible that the
inconsistent responses are caused by bias. As a result, the researcher needs to edit
the contradictory responses provided by the respondents. Meanwhile, illegal codes
are values that are not indicated in the coding instructions provided.
Chapter 3- Research Methodology 46

3.6.3 Data Transformation

Additionally, the researcher also carried out data alteration after data entry
and editing. Data transformation is a data coding variation, which is the process of
altering the original numerical representation of a quantitative value to another value
(Sekaran and Bougie, 2010). The data transformation that had been implemented in
this research was reverse scoring. For instance, almost all the questions are in
positive forms and there are only four questions, which are under training part
(independent variable) and Question 6 under Job involvement part are in negative
form. It means that most of the responses in question will be somewhere in scale 1
(Strongly Disagree), 2 (Disagree), or 3 (Slightly Disagree). Thus, to maintain
consistency in the meaning of response, we reversed the score for these questions.
For example, a 1 (Strongly Disagree) would be transformed to a 7 (Strongly Agree),
a 2 to a 6, and so forth.

3.7 Data Analysis

According to Zikmund et al. (2010), data analysis is defined as the reasoning
application which helps the researchers to understand the data that have been
collected. The purpose to implement data analysis is to examine and model the data
by assigning facts and figure to answer research problem. Also, it highlights the
useful information by recommending assumptions to take advantage of the collected
data in order to solve some specific problem, such as addressing the research
problem.

The computer software that has been applied to analyse the collected data is
SPSS. SPSS provides many types of analysis that is very helpful in current research.
Typically, there are three types of analysis that is required in this research, which are
descriptive analysis, scale measurement, and inferential analysis.

3.7.1 Descriptive Analysis

Descriptive analysis is used in order to elucidate and describe the
characteristics of the variables of interest in a situation (Sekaran and Bougie, 2010).
Besides, Zikmund defined descriptive analysis as the basic alteration of data in a
way that illustrate the fundamental characteristic, such as central tendency and
Chapter 3- Research Methodology 47

variability. Generally, mean, median, mode, variance, range, and standard deviation
are widely applied in describing the descriptive statistic. The advantage of using
descriptive analysis is that it helps to summaries the sample and measure. It also
forms basic quantitative data analysis with simple graphics analysis.

In this research, the researcher choose to use pie chart for analysing the
collected data. A pie chart is circular chart in which the circle is cut into pieces or
sectors. Each sector in the circle represents an item in a data set to match the amount
of item.

The pie chart is undertaken to determine the percentage of type of

employment, years of working experience, gender, age, race, marital status, and
level of education. Meanwhile, pie chart is easily for us to differentiate the sizes of
items in one chart. Also, it requires less additional explanation comparing to others
types of graphics (such as histograms). Indeed, the summary of the overall data set
in visual form are easily to understand by every researcher.

3.7.2 Scale Measurement- Reliability Test

In order to obtain consistent result, it is critically important to ensure that all
the questions in the survey form are clear and free from ambiguity and error before
the actual survey has been carried out. That is why reliability test is needed to be run
when pilot study is done. The reliability is normally measured by the Cronbach’s
Alpha.

According to Zikmund et al. (2010), the Cronbach’s Coefficient Alpha is

ranging in value from “0”, meaning no consistency, to 1, meaning complete
consistency (all items produce corresponding values). The independent variables and
dependent variable that are required to run the reliability test are motivation,
personal factors, training, job characteristics, and job involvement. In fact, all the
alpha value for each independent variable must be above 0.70 which reveals that the
recommended independent variables are good or sustained reliable (Jagannathan,
2008).
Chapter 3- Research Methodology 48

Table 3.2: Rule of Thumb of Cronbach’s Coefficient Alpha

Coefficient Alpha Range , Į Description

0.80 to 0.95 Very Good Reliability
0.70 to 0.80 Good Reliability
0.60 to 0.70 Fair Reliability
Į Poor Reliability
Source: Zikmund, W. G., Babin, B. J., Carr, J. C., & Griffin, M. (2010). Business Research the
Methods (8th ed.). New York: South-Western/ Cengage Learning.

3.7.3 Inferential Analysis

Pearson Correlation Coefficient and Multiple Regression Analysis are being
selected in interferential analysis. The methods are used to analyse the relationship
between independent variables (Motivation, Personal Factors, Training, and Job
Characteristics) and dependent variable (Job Involvement).

3.7.3.1 Pearson Correlation Coefficient

According to Sekaran and Bougie (2010), Pearson correlation coefficient
will provide the researchers with the direction, strength and significance of the
bivariate relationship among all the variables that are measured through the use of
interval or ratio scale. The condition for using this correlation is to test the
relationship between independent variable and dependent variable. It is also
considered as a powerful test which always gives credible result to the researchers.
The correlation is derived by evaluating the changes in one variable as another
variable also varies.

In theory, Pearson correlation coefficient can be divided into two

correlations, one is perfect positive correlation between two variables which is
always represented by 1.0 (plus 1.0), and another is perfect negative correlation
which is always represented by -1.0 (minus 1.0). Nevertheless, Sekaran and Bougie
(2010) stated that neither of these correlations will be discovered in reality when
evaluating correlations between any two variables expected to be dissimilar from
each other.
Chapter 3- Research Methodology 49

The reason that the researcher choose this statistical test is that correlation
coefficient is suitable for measuring the independent variables and dependent
variable which are designed in Likert scale.

Table 3.3: Rule of thumb of Pearson Correlation Coefficient

Coefficient Range Strength

+/- 0.91 to +/- 1.00 Very strong
+/- 0.71 to +/- 0.90 High
+/- 0.41 to +/- 0.70 Moderate
+/- 0.21 to +/- 0.40 Small but definite relationship
0.00 to +/- 0.20 Slight, almost negligible
Source: Hair, J. F., Money, A. H., Samouel, P., Page, M. (2007). Research methods for business.
Chichester, West Sussex: John Wiley & Sons, Inc.

3.7.3.2 Multiple Regressions

Multiple regression analysis is similar to simple regression analysis. The
difference between these two analyses is that simple regression just tests on the
impact of one independent variable towards one dependent variable, while multiple
regressions are used to test the impacts of more than one independent variable
towards one dependent variable. Meanwhile, all the independent variables and
dependent variable must be in the form of Likert scale. As a result, we applied this
to examine relationship among all the independent variables with the dependent
variable. With this analysis, it helps us to clearly identify the most influential factor
(independent variable) that affects employees’ job involvement (dependent
variable). We can identify the importance of independent variables by comparing the
beta weights.

3.8 Hypothesis
In order to justify the above given objectives following hypothesis were
framed, which will be tested in the further sections of this study on certain
parameters of inferential statistics. In order to achieve the above said objectives,
following hypothesis will be tested.
(1) Highly motivated employees are more efficient than less motivated
employees.
Chapter 3- Research Methodology 50

(2) There is significant relationship between motivation and job involvement.

(3) There is significant relationship between personal factors and job
involvement.

3.9 Conclusion
In this chapter, it discusses about the methodology function and purposes as
well as the way the research was conducted where the researcher used questionnaire
as primary data and journal as secondary data. Also, the researcher had done a pilot
test study and reliability test based on the data collected from the pilot study.
Besides, we also explained the functions of Statistical Package of the Social
Sciences (SPSS), Cronbach Coefficient Alpha, Pearson Correlation Coefficient, and
Multiple Regressions in this chapter. Thus, in the following chapter, the researcher
will explain in detailed about systematic understanding by illustrating and
interpreting the data collected from the actual survey and pilot test.