44 CHAPTER-3 RESEARCH METHODOLOGY 3.1 INTRODUCTION This chapter shall elaborate on the methodology adopted by the researcher to conduct the proposed study. This chapter shall throw light on the research design adopted, nature of data collected, sources of data, the sampling plan proposed to be used, the research instrument to be utilized for the research and details relating to the representation and analysis of the collected data. 3.2 RESEARCH DESIGN The proposed research study is descriptive in nature, covering manufacturing industries situated in Union Territory of Puducherry. 3.3 NATURE AND SOURCE OF DATA Both primary and secondary data have been used for this research. Primary data was collected using a well structured questionnaire, which was administered personally to the executives of manufacturing undertakings in Union Territory of Puducherry. Secondary data was collected from the findings of Published Papers, Articles, Books, Prior Studies, Organizations‘ Bulletins, Annual Reports of the manufacturing units and from various web sites. 3.4 DATA COLLECTION INSTRUMENT 3.4.1 Initial Items Generation The survey instrument was initially developed based on review of previous literature, which addressed the basic theoretical constructs of Business Environment Characteristics of Manufacturing undertakings, Advanced Manufacturing Technologies of Manufacturing undertakings, Competitive Priorities of Manufacturing undertakings and Business Performance of Manufacturing undertakings. These constructs were further sub-divided into various domains, each consisting of number of statements. The Business Environment Characteristics construct is divided into six domains namely, Labour Availability, Business cost, Competitive Hostilities, Dynamism, Political Environment and Government Laws and Regulations. Similarly, the Advanced Manufacturing Technologies construct is sub-divided into three domains namely, Advanced Manufacturing Technology Implementation, Direct Advanced Manufacturing
30
Embed
CHAPTER-3 RESEARCH METHODOLOGY 3.1 INTRODUCTION 3.2 ...shodhganga.inflibnet.ac.in/bitstream/10603/23402/12/12_chapter 3.p… · RESEARCH METHODOLOGY 3.1 INTRODUCTION This chapter
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
44
CHAPTER-3
RESEARCH METHODOLOGY
3.1 INTRODUCTION
This chapter shall elaborate on the methodology adopted by the researcher to
conduct the proposed study. This chapter shall throw light on the research design
adopted, nature of data collected, sources of data, the sampling plan proposed to be used,
the research instrument to be utilized for the research and details relating to the
representation and analysis of the collected data.
3.2 RESEARCH DESIGN
The proposed research study is descriptive in nature, covering manufacturing
industries situated in Union Territory of Puducherry.
3.3 NATURE AND SOURCE OF DATA
Both primary and secondary data have been used for this research. Primary data
was collected using a well structured questionnaire, which was administered personally to
the executives of manufacturing undertakings in Union Territory of Puducherry.
Secondary data was collected from the findings of Published Papers, Articles, Books,
Prior Studies, Organizations‘ Bulletins, Annual Reports of the manufacturing units and
from various web sites.
3.4 DATA COLLECTION INSTRUMENT
3.4.1 Initial Items Generation
The survey instrument was initially developed based on review of previous
literature, which addressed the basic theoretical constructs of Business Environment
Characteristics of Manufacturing undertakings, Advanced Manufacturing Technologies
of Manufacturing undertakings, Competitive Priorities of Manufacturing undertakings
and Business Performance of Manufacturing undertakings. These constructs were further
sub-divided into various domains, each consisting of number of statements.
The Business Environment Characteristics construct is divided into six domains
namely, Labour Availability, Business cost, Competitive Hostilities, Dynamism, Political
Environment and Government Laws and Regulations. Similarly, the Advanced
Manufacturing Technologies construct is sub-divided into three domains namely,
Advanced Manufacturing Technology Implementation, Direct Advanced Manufacturing
45
Technology and Indirect Advanced Manufacturing Technology. The third construct of
Competitive Priorities is divided into six domains namely, Quality, Cost, Delivery,
Flexibility, Customer Focus and Knowhow. The fourth and final construct of Business
Performance consists of five statements. Totally 109 variables were included in the
preliminary schedule, of which 37 were included in the first construct, 24 in the second
construct, 43 in the third construct, and 5 variables were included in the fourth and final
construct.
The executives of the manufacturing firms were asked to provide perceptual
information on the performance of their company. Dess and Robinson (1984) have
recommended the utilization of perceived measures if objective measures are not
available. It is difficult to collect financial data of manufacturing undertakings from the
executives as the data may be confidential. Further, the executives may not have the data
in their memory and may not be able to recall the desired data for the study when the
survey is undertaken. Swamidass and Newel (1987) have also confirmed the difficulties
of using objective measures of performance due to the reluctance of the companies to
provide financial data. Hence, the executives of manufacturing units have been required
to rate the comparative market share and growth of sales of their undertaking in relation
to their competitors, in a Likert‘s Five Point Scale. Vickery et al. (1993) has also used
perceptual information to study the trends of Return of Investment and Return on Sales of
undertakings. Hence, the researcher decided to use the perceived measurement of
performance of the manufacturing units.
3.4.2 Qualitative Inquiry
Researcher held extensive consultations with the research supervisor and other
subject experts and industrial experts in the field of Operations Management. An In-depth
interview was conducted with a panel of ten best academicians, consisting of four experts
from Operations Management, two Statisticians, two Finance experts, one English expert
and one representative of the State Government. These experts from top and eminent
institutions such as Pondicherry University, NIT Tiruchy, and Anna University, Chennai.
Furthermore, the researcher held extensive consultations with ten industrial experts in the
rank of Chief General Managers and Managers at operational level and got valuable
inputs in drafting the schedule.
46
3.4.3 Face Validity and Content Analysis
The experts consulted were requested to carefully go through the research
constructs in the light of the objectives of the research. The experts evaluated the 109
constructs and selected the constructs to be included for the study based on the
compatibility and, representativeness, suitability and importance of the constructs, and
the capacity of the constructs to match with domain area and yield best accurate results,
those constructs not satisfying these parameters shall be summarily rejected and removed
from the schedule. Further, additional constructs suggested by the experts shall be
included in the schedule. Furthermore, the experts were requested to evaluate the
complete schedule for its simplicity, clarity, unambiguity, composition of the schedule,
feasibility of obtaining the desired information from the respondents and the length of the
schedule, and accordingly refine the schedule. The refined scale was further improved
based on the suggestions of industrial experts. Upon the successful completion of this
process, the number of variables constituting the Business Environment Characteristics of
Manufacturing firms were reduced to five for Business cost domain, five for Labour
Availability domain, five for Competitive Hostilities domain, four in respect of
Government Laws and Regulations domain, five in respect of Political Environment and
four in respect of Dynamism. Similarly, the number of variables constituting the
Advanced Manufacturing Technologies construct is arrived at six for Advanced
Manufacturing Technology Implementation domain, six for Direct Advanced
Manufacturing Technology domain and five in respect of the Indirect Advanced
Manufacturing Technology domain. The number of variables constituting the third
construct of Competitive Priorities is fixed at six for Quality domain, five for Cost
domain, five for Delivery domain, four for Flexibility domain, four for Customer Focus
and six in respect of the Knowhow domain. The fourth and final construct of Business
Performance consists of five variables without any sub-divisions of domains.
Hence, the total number of variables pertaining to Business Environment
Characteristics was reduced to 28, while those relating to Advanced Manufacturing
Technologies were reduced to 17, those relevant to Competitive Priorities were reduced
to 30 and the number of variables relating to Business Performance of the Manufacturing
undertakings was arrived at 5. Hence, the total numbers of constructs were reduced from
47
109 to 80. Further, the number of questions relating to the industrial profile of the
manufacturing units studied was enhanced from 10 to 13.
3.4.4 Pilot Study
Prior to the full-fledged resumption of the research process, a pilot study was
conducted on some 52 manufacturing undertakings located in Puducherry. Based on the
feedback obtained from the Pilot study, the researcher made minor modifications in the
questions pertaining to the industrial profile of the manufacturing units studied. These
questions were redesigned in statement forms to accommodate the respondent‘s
recommendations. Further, based on their feedback, some technical terms which were not
easily understandable for the respondents were suitably modified and substituted with
simpler terms.
Based on the inputs obtained from the Pilot study, the final schedule was drafted.
The final schedule consisted of five sections. The first section consisted of 13 questions
relating to the industrial profile of the units studied, while the second section consisted of
28 key determining variables to measure the business environment characteristics of the
manufacturing units. The third section endeavours to measure the advanced
manufacturing technologies of the manufacturing undertakings with the help of 17
variables in respect of 3 domains, while the fourth section tries to measure the
competitive priorities of the manufacturing units through 30 variables. The fifth and final
section consists of 5 key determining variables to measure the business performance of
the manufacturing undertakings. Five point Likert‘s Scale has been used in respect of
these 80 variables, with the scale values ranging from low level of priority attached to
high level of priority.
3.4.5 Reliability test for Data Collection Instrument
The next step is to test the reliability of the schedule. Reliability shall reveal the
accuracy and consistency of the results from the survey instrument. The result of the Pilot
study was tested for Reliability using the Cronbach alpha and the value of reliability was
found to be more than satisfactory level of 0.6 in respect of all the categories.
In addition to testing the reliability, the researcher has also tested the
Communality, which measures the percent of variance explained by the factors in a given
item. Furthermore, the researcher has tested the Normality, which indicates the normal
48
distribution of the data. While plotting the data in a graph, if a bell shaped curve is
arrived at, then the mean value is 0 and the value of standard deviation is 1, which
indicates that there is a standard normal distribution of the data (Lewis-Beck, Bryman &
Liao, 2004; Groebner & Shannon, 1990). Testing Normality is absolutely important for
multivariate data analysis (Hair et al. 2006).
The researcher then proceeded to test the Homogeneity of the data, which shall
indicate the uniqueness of the population. In addition, the researcher proceeded to test the
Multicollinearity, which presence when more than two independent variables represent
the common thing. Tabachnick & Fidell (2007) in his study suggested items from the
same construct in the data set, the correlation value higher than 0.90 between any
variables will create some problematic conditions in analysis. If it is more than above
author suggested value it is better to exclude a particular item is advisable. To evaluate
Multicollinearity, item to item correlations were calculated between each item. It helps to
solve the multicollinearuty problem in the data set.
Next, the researcher tested the Linearity of data, which reveals the existence of
linear relationships among the variables, which is important in multivariate analysis
techniques. Most multivariate techniques (including covariance structure modeling)
employed in this study implicitly believe that relationships between variables are linear.
Departures from linearity have an effect on calculated correlations between variables in
this study were carefully evaluated the linearity. The next step is to test the Individual
item reliability using factor loading to analyse the individual reliability of the variables.
Carmines and Zeller (1979) proposed factor loadings greater than or equal to 0.707.
However, several authors (Barclay et al., 1995; Chin, 1998) recommended that this
principle not to be supposed to strictly followed in exploratory studies; in some
circumstances factor loadings up to 0.5 or 0.6 can be accepted.
The next step is to test the Construct reliability, which sure that the internal
consistency of all the variables when they compute the similar idea by assessing how
aggressively the observable variables evaluate the latent variable (Fornell and Larcker,
1981). Construct reliability value should be more than 0.6 and it shows acceptable
reliability of the measurement items (Chen and Paulraj, 2004; Nunnally, 1978; Cronbach,
1951). Further, the researcher has proceeded to test the Convergent validity, which
49
establishes the relationship between observed constructors. It is the process of analyzing
scores of one constructor correlated with the scores of another measure, may be similar or
different. Fornell and Larcker (1981) assessed convergent validity, by using the average
variance extracted (AVE) and value should be 0.5 or above. The researcher has also
tested the Discriminant validity, which specifies that how much the given construct with
another construct in a model. This validity can be calculated by matching the AVE value
with the square of the correlations of the constructs.
3.5 SAMPLING
3.5.1 Sample Frame
Union Territory of Puducherry is the sample frame for the study. All the four
regions of Union Territory namely, Puducherry, Karaikal, Yanam, and Mahe. With a
current population of 11.1 lakhs and existence of well established 72 large scale
industries, 176 medium scale industries and 7950 small scale industries and the this
number being on the ever increase, offers tremendous scope for choosing Puducherry as
the sample frame for the study.
3.5.2 Sample Population
Business units engaged in manufacturing and those located in the four regions of
Pondicherry, Karaikal, Mahe and Yanam shall be the sample population for this study.
executives with titles of Directors, Chief Executives, Managing Directors, General
Managers and Senior Level Managers who have the leadership in different functional
areas like Operations, Marketing, Human Resources and Finance shall constitute the
sample population, from which sample shall be drawn for the conduct of this study.
3.5.3 Sample technique
The sample technique used for the study is Simple Random sampling method.
The names of 8588 units engaged in manufacturing as on 2011 were listed out and 365
sample units were drawn from this list using the Lottery Method.
3.5.4 Sample size
The most important part of any research is the proper calculation of appropriate
sample size for the survey. In the present research work, the formula
205.096.1 n has been used to calculate the appropriate sample size for the
study. Pilot study was conducted on manufacturing firms located in Puducherry. Based
50
on the findings of the pilot survey, a possible sample size of 337 with a 5% error in mean
estimate was finalized for the study. Out of 365 questionnaires which were administered,
15 were rejected for invalid and incomplete responses and 350 valid questionnaires were
considered for further analysis which is more than the desired figure of 337.
3.6 DATA COLLECTION METHOD
Personal Interview method was employed to collect data. The researcher
administered the schedule personally to the respondents and collected the necessary data.
3.7 DATA REPRESENTATION
The raw data collected were coded suitably and represented in tabular and
diagrammatic forms to facilitate the usage of traditional and sophisticated statistical tools
for analyzing the data.
3.8 DATA ANALYSIS TOOLS
Both traditional and sophisticated statistical tools were applied for data analysis.
The data collected were fed in to Excel sheet and the statistical packages of SPSS 19
Version and LISREL were employed. The statistical tools of Mean, Standard Deviation,