Business Research

Methods
Project

Group 4

Submitted to: Prof. Alok Kumar

Submitted by: ​Ayushi Roy (143067)

Deepa M(143068)
Gauransh(143072)
Ishan Kumar(143073)
Omita Agarwal(143084)
Rachana G Krishnan(143089)
Rutba Mufti(143093)
Vishnu Parvanda(143113)
Vivek Kumar Pandey(143115)
Akash Jain (143120)
 Table of Contents

S No Title Page No
1
2
3
4
5

INTRODUCTION TO FACTOR ANALYSIS

Factor analysis is a method used to simplify a large number of variables into smaller factors.
Maximum common variance is extracted from all the variables and put together into a common
score. This score is further used for analysis, as an index of all variables. Factor analysis is a part
of the general linear model and has certain assumptions which are as follows:
· There is a linear relationship
· There exists no multicollinearity
· Relevant variables are included into analysis
· True correlation between factors and variables exists
Types of factoring:
In order to extract the factor from a data set, different types of methods are used. Some of them
are as follows:
1. Principal component analysis: This is the most frequently used method by
researchers. PCA begins extracting the maximum variance and ten punts them into
the first factor. Then, the variance explained by the first factors are removed and
maximum variance is extracted from the second factor. This process is continued till
the last factor.
2. Common factor analysis: This is the second most frequent method used by
researchers. Under this, the common variance is extracted and put into factors.
However, unique variance of all variables is not included in this method. In Standard
error of mean (SEM), this method is used.
3. Image factoring: In correlation matrix, this method is used. In order to predict the
factor in image factoring, OLS (Ordinary least squares) regression method is used.
4. Maximum likelihood method: This method works on correlation metric as well,
however, it uses maximum likelihood method to factor.
5. Other methods of factor analysis: Least squared is outweighed by alfa factoring.
Another method used for factoring is weight square based on regression.
​Factor loading:​ It is the coefficient of correlation for the factor and variable. It shows the
variance that is explained by the variable on that particular factor. As a rule of thumb in the SEM
approach, 0.7 or higher factor loading represents that sufficient variance from that variable has
been extracted by the factor.
Eigenvalues:​ It is also known as characteristic roots. Out of the total variance, the variance
explained by that particular factor is shown in the eigenvalues. The commonality column shows
as to how much variance is explained by the first factor out of the total variance. For example, if
the first factor explains 72% variance out of the total, this means that 28% variance will be
explained by the other factor.
Factor score:​ It is also known as the component score. This score is of all rows and columns,
which can be used for further study as an index of all variables. By multiplying a common term,
we can standardize this score. Whatever analysis we do, with this factor score, we will assume
that all variables behave as factor scores and will shift.
Criteria for determining the number of factors:​ Eigenvalues is a good criterion for
determining factor, according to the Kaiser Criterion. We should consider it a factor if the
Eigenvalues is greater than one and vice versa. It should be greater than 0.7 according to the
variance extraction rule. If the variance is less than 0.7, it shouldn’t be considered as factor.
Rotation method:​ This method makes it more reliable to understand the output. Rotation
method affects the Eigenvalues or percentage of variance extracted even though Eigenvalues do
not affect the rotation method. The rotation methods available are as follows:
· No rotation method
· Promax rotation method
· Direct oblimin rotation method
· Varimax rotation method
· Quartimax rotation method
All of the above can be easily selected in SPSS and our variance explained by those particular
methods can be compared.

Reliability Testing

Reliability is described as the probability of failure-free software operation for a specified

period of time in a particular environment.
Reliability testing is carried out to ensure that the programme is stable, fulfils the function for
which it is made, and is capable of conducting a fault-free operation for a specified period of
time in a given environment. In today's world, in every area of our lives, software applications
are used, including healthcare, government sectors, telecommunications, etc.​ ​Therefore, we need
to provide reliable information that users can rely on. The consistency of the applications and the
standardization of products deal with reliability testing. If we can replicate the test cases and if
we consistently get the same result, then the product is said to be 'reliable'.
The type of reliability you should calculate depends on the type of research and your
methodology.
Factor Analysis Output Interpretation:

To continue with factor analysis we first check all the eligibility criteria for factor analysis.

Eligibility criteria for Factor Analysis:

1. Respondents data should be at least 4 to 5 times the number of variables
2. KMO value should be more than 0.5
3. Significant correlation among variables.

The data set passes the first eligibility criteria (Respondents almost 200 > 5 times 39).

To check the second eligibility criteria look at the KMO value of data set in the table “ KMO and
Bartlett’s test”
It is ​.874​, since it is more than 0.5 the data set passes the second eligibility criteria.

To check the third eligibility criteria, develop a null hypothesis.

Ho= Correlation among variables is insignificant or correlation matrix is an identity matrix.

In the table “ KMO and Bartlett’s test” look at the ​Sig. value. ​Since it is less than the alpha value
(0.05) we reject our null hypothesis. Hence, we can conclude that there is significant correlation
among the variables and our data set passes all three criteria for the factor analysis.

Therefore, it passes all the three eligibility criteria.

Rotated Component Matrix Table:

From this table we develop constructs or factors. Cut off value for variable is 0.5
Now we club different variables into various components and then name them.

Component Number Variables clubbed together

1 21,24,25,27,28,30,35,37,38,39

2
1,2,9,14,15,16,17,22,26

3
6,18,19,20,28
 4
31,32,33

5
11,12

6
5,10

7
23,34,36

8 8

9
29,33

Variables which were not used to make factors: ​3,4​ because didn’t clear the cut off
point.

After clubbing the factors we name them.

 Questions Factor Name

Factor 1 Q2.21Green products Intention to recommend

make my life more about
protecting the environment

Q2.24 I like the Green

products outlook

Q2.25 My friends would

have a positive opinion
when they discovered I
purchased

Q2.27 Using Green

products makes me feel
relaxed and happy

Q2.28Green products are

good for the environment

Q2.30 Green products can

effectively reduce pollution

Q2.35I intend to buy the

Green Products out of
concern for the
environment

Q2.37I would recommend

the Green Products to my
friends and family to use it

Q2.38If I have a worthy

experience with the Green
Products I would
recommend
 Q2.39I would recommend
the Green Products if it is
worth using

Factor 2 Q2.1Green products Perceived ease of use

represent comfort

Q2.2Green products are of

high quality

Q2.9 Green products are

family oriented

Q2.14 Quality of the Green

products are good

Q2.15Green products have

fully quality guarantees

Q2.16 Design of the Green

products are impressive

Q2.17 Function of the

Green products fit my
needs

Q2.22I think Green

products are more reliable
than non green products

Q2.26 The function of the

Green products is
trustworthy
Factor 3 Q2.6 Green products are Eco friendliness
low air polluting

Q2.18 Green products are

made of recyclable
materials

Q2.19 Green products are

associated with protecting
the environment

Q2.20 Green products

inspire environmental
debates

Factor 4 Q2.31 Green products are Perceived cost

expensive

Q2.32Green products cost

more than non green
products

Q2.33Green products are

cheaper than non green
products

Factor 5 Q2.11 Green products are Trust

respected

Q2.12Green products are

friendly
Factor 6 Q2.5 Green products are Awareness
associated with high
technology

Q2.10 Green products are

well known

Factor 7 Q2.23 I think Green Self affirmation

products can represent my
socioeconomic status

Q2.34 I intend to buy the

Green Products in order to
fulfil my needs

Q2.36 The possibility of my

purchasing Green
Products is high

Factor 8 Q2.8 Green products are Creativity

creative

Factor 9 Q2.29Green products Perceived value

cannot help slow the
deterioration of the
environment

Q2.33Green products are

cheaper than non green
products

Total Variances Explained

a. ​Factor​ – The initial number of factors is the same as the number of variables used in the
factor analysis. However, not all 39 factors will be retained. In this case, only the first nine
factors will be retained.

b. ​Initial Eigenvalues​ – Eigenvalues are the variances of the factors. Because we conducted
our factor analysis on the correlation matrix, the variables are standardized, which means that
each variable has a variance of 1, and the total variance is equal to the number of variables used
in the analysis.

c. ​Total​ – This column contains the eigenvalues. The first factor will always account for the
most variance (and hence have the highest eigenvalue), and the next factor will account for as
much of the left over variance as it can, and so on. Hence, each successive factor will account
for less and less variance.

d. ​% of Variance​ – This column contains the percent of total variance accounted for by each
factor.

e. ​Cumulative %​ – This column contains the cumulative percentage of variance accounted for
by the current and all preceding factors.

f. ​Extraction Sums of Squared Loadings​ – The number of rows in this panel of the table
correspond to the number of factors retained. In this example, nine factors are retained, so there
are nine rows, one for each retained factor. The values in this panel of the table are calculated in
the same way as the values in the left panel, except that here the values are based on the common
variance. The values in this panel of the table will always be lower than the values in the left
panel of the table, because they are based on the common variance, which is always smaller than
the total variance.

g. ​Rotation Sums of Squared Loadings​ – The values in this panel of the table represent the
distribution of the variance after the varimax rotation. Varimax rotation tries to maximize the
variance of each of the factors, so the total amount of variance accounted for is redistributed over
the nine extracted factors.

In our case, nine factors which have Eigenvalues more than one have been chosen from the
Principal Component Analysis.

Reliability Test

Conditions for reliability test:

To test the reliability of the questionnaire, Cronbach's alpha value is calculated to measure the
internal consistency of the questionnaire, of a set of scale or test items. Cronbach's alpha is thus a
function of the number of items in a test, the average covariance between pairs of items, and the
variance of the total score.

The condition for the reliability test to be acceptable is when the Chronbach’s alpha value is
more than 0.7
When the Chronbach’s alpha value is less than 0.7 it shows that the questionnaire is not
internally consistent

Reliability test output:

FACTORS Alpha Coefficient Accepted/ Rejected

1.​ A
​ wareness 0.542 Rejected

2.​ E
​ co-friendliness 0.774 Accepted

3.​ I​ ntention to Recommend 0.887 Accepted

 4.​ P
​ erceived Cost -0.70 Violates reliability model
assumptions

5.​ P
​ erceived Ease of Use 0.882 Accepted

6.​ P
​ erceived Value 0.073 Rejected

7.​ S
​ elf-Affirmation 0.696 Rejected

8.​ T
​ rust 0.542 Rejected

Total Questionnaire Set 0.932 Accepted

Observation:​ Looking at the results we can infer that there is considerable consistency between
the questions in the entire questionnaire as the Cronbach alpha value is 0.932. Hence, this
questionnaire can be used to find the results of the research accurately.