Professional Documents
Culture Documents
Chapter 3 29.12.22
Chapter 3 29.12.22
Research Design
3.1 Introduction
In this study the research methodology was adopted to examine the factors of EI with
reference to IT employees and to empirically evaluate the strength of each factor. The
research design that was adopted was descriptive in nature as quantitative analysis has been
adopted for the study. The opinions and behaviours of the respondents were measured using
quantitative research. To simplify the data generated to comprehend the significance between
independent and dependent variable a statistical, mathematical or computational techniques
were used.
SPSS version 20 and AMOS were used for carrying out the data analysis (Arbuckle, 2017).
The responses were collected from the faculty members employed in the higher education
sector through the self-reported method. “Well-Structured, validated and reliable
questionnaire was used to carry out the survey.” Descriptive statistics was done for
computing the Means and Standard Deviation (SD), Reliability, Normality, Exploratory
Analysis Factor, ANOVA, linear regression, test for significance was taken to study the
constructs. Path analysis was done using Structural Equation Model (SEM).
Main Study: The researcher divided Bangalore into North, South, East and West zone and
identified IT firms in each of these zones in order to collect the data from the IT employees
working there. Around 423 responses were collected from the IT employees working in 20 IT
firms over a period of 18 months.
Sample Size determination
The sample size was determined using Cochran’s formula for sample size determination.”
Cochran’s formula is normally used in situations with large population. The Cochran’s
formula is:
Sample Size = Z2pq/e2
Where e = margin of error
p = population proportion in question
q=1–p
The Z value can be found in the Z table.
In this study:
p = 0.5
The confidence level considered in this study is 95%, hence the margin of error would be
e = 0.05
In the Z table, the value for Z stands at 1.96 at 95% level of confidence.
Therefore,
“Sample size = (1.96)2(0.5)(0.5) / (0.05)2
Sample size = 385 IT employees
To account for individuals that the researcher is unable to reach, often researchers raise the
sample size by 10%. To account for nonresponse, the sample size is often increased by 10%.
Hence the revised sample size = 423 IT employees.
Self-Confidence 4 0.908
Emotional Self-
4 0.837
Control
Transparency 4 0.735
Self- Adaptability 4 0.779
Management
Achievement 4 0.813
Initiative 4 0.811
Optimism 4 0.911
Developing
4 0.769
Others
Inspirational
4 0.853
Leadership
Conflict
4 0.849
Management
Teamwork &
4 0.839
Collaboration
The above table shows the ECI 2.0 tool used for the study along with their sub-dimensions
and the number of items in the measuring instrument that measures each of those factors. The
above table also shows the Cronbach alpha values for each sub-dimension scale and items
deleted (if any) from each scale.
KMO and Bartlett's Test – Self-Awareness, Self-Management, Social Awareness &
Relationship Management
“KMO and Bartlett's Test” SA SM Soc. A RM
From the above table, it is clearly seen that the KMO and Bartlett’s Test of Sphericity is
significant indicating that the sample is adequate as well as indicating that the Principal
Component Analysis (PCA) could be undertaken.
Phase 1: “Pilot Study”
It was conducted to analyse the study‘s viability, time, budget, and statistical erraticism. It
provides an opportunity to improvise the research methodology toward executing the final
study. This will provide useful insights to the investigator to develop the study into a robust
and fruitful exercise. A structured and valid questionnaire of seventy two (72) items of ECI
2.0 tool with 10 demographic questions in five (5) sections was used for the pilot study
survey.
First section: “This section comprised of 10 items corresponding to the respondent’s
personal details; Name (optional), Gender, Age, Marital Status, Qualification, Annual
Income, Work experience, Job Role & Name of the Organization.”
Second section: This section comprised of 12 items of the Self-Awareness sub-scale of ECI
2.0
Third section: This section comprised24 items of the Self-Management sub-scale of ECI 2.0
Fourth section: This section comprised of 12 items of the Social Awareness sub-scale of ECI
2.0
Fifth section: This section comprised of 24 items of the Relationship Management sub-scale
of ECI 2.0
Reliability Test: It measures the consistencies of the measuring instrument over a period of
time and between different respondents to produce consistent results. The study adopted
reliability test to analyse the construct items for Cronbach coefficient alpha. Similarly the
internal stability of the items: individual, sub-dimension and total construct items were
analysed. The study accepted Coefficient alpha beyond (0.70) for the construct scale or items.
Normality Test: This test was used to determine if the data set was well-modelled and
normally distributed. The measure of Skewness and Kurtosis indicated the normalcy of data.
Histograms indicated the data is normally distributed.
Descriptive Statistics: This technique was used to measure the validity of descriptive
coefficients that summarise the data representation of the sample population. It is also used to
calculate the intermediate propensity and (spread) variability. Distributions of data of the
construct were measured using mean and the standard deviation.
Regression Analysis: In the statistical method, linear regression is adopted to examine the
relationship and its influence between independent variables is related to dependent variables.
Also establish the types of these relationships.
Tests for significance of mean differences: This method is engaged to determine the
difference between construct sample mean and hypothetical mean including population
variables. It is measured using ‘t-test and One-way ANOVA’. Further, post-hoc tests were
used to exposes the difference amid group means which has three or more clusters using
(ANOVA) F test of significance.
Structure Equation Model (SEM): It is based on diverse models and statistical approach
that fits the construct relationships of the research data. It is a blend of analysis that is
regression, factor or path.