You are on page 1of 20

Data Analysis & Interpretation

Demographic variable Frequency Total


Male 100
Gender
Female 0
Graduate 1
Higher secondary 8
Qualification Secondary 30
Primary 42
Illiterate 19
Below 10000 46
10000 – 25000 48 100
Income 25001 – 40000 2
40001 – 60000 3
Above 60000 1
18 to 30 years 8
31 to 40 years 20
Age 41 to 50 years 32
51 to 60 years 3
Above 60 years 3

Demographic variable Frequency Total


Below 5 Year 12
Experience of 5 – 10 years 12
driving rickshaw 10 – 15 years 28
Above 15 years 48
100
Below 5 Year 9
Experience of 5 – 10 years 15
Owning rickshaw 10 – 15 years 29
Above 15 years 47
1–5 92
Number of
5 – 10 4
rickshaw own 100
10 – 15 2
Above 15 2
1 – 50 km 1
How much do you 50 – 100 km 1
drive during one 100 – 150 km 8 95
day 150 – 200 km 39
Above 200 km 46
Have you heard I have not heard about it 16
about solar e- I have heard bit about it 67 100
rickshaw I have heard a lot about it 17
Have you ever No, I have not 21
seen solar e- Yes, I have 75
rickshaw in real life Don’t remember 4

If you have heard about solar e rickshaw, then how much do you agree
with following statement

SA A N DA SDA Total
A solar e-rickshaw is more reliable 28 36 11 6 4 85
A solar e-rickshaw accelerates slower 29 26 22 4 4

A solar e-rickshaw emits less pollution 67 14 3 0 1

Solar e-rickshaws have unique features 19 42 18 6 0

Solar e-rickshaws are much cheaper to own 53 21 5 4 2

Solar e-rickshaw doesn’t offer enough 24 41 12 6 2


performance and pickup
Gas prices are not high enough to justify 10 28 37 8 2
buying a solar e-rickshaw
Chart Title
80
70
60
50
40
30
20
10
0
A solar e- A solar e- A solar e- Solar e- Solar e- Solar e- Gas prices are
rickshaw is rickshaw rickshaw rickshaws rickshaws are rickshaw not high
more reliable accelerates emits less have unique much cheaper doesn’t offer enough to
slower pollution features to own enough justify buying
performance a solar e-
and pickup rickshaw

SA A N DA SDA

ant factor
Predomin
Not at all

A Little

Some

Total
what

A lot

Purchase price 4 5 13 54 24

Mileage 0 9 21 33 37
Maintenance Cost 0 5 33 31 31

Resell Value 2 6 43 24 25
Speed 0 4 17 51 28

Total range after refueling 1 13 52 24 10 100


Pollution 14 15 20 34 17

Power/Pickup 0 5 12 39 44

Manufacturer Advertisement 3 26 52 15 4

Observation from friends and family 5 12 22 31 30


Proximity of dealership for service 20 23 30 16 11
Chart Title
60
50
40
30
20
10
0

Not at all A Little Some what A lot Predominant factor

Petrol Gas Solar E-rickshaw Total


The most preferred alternative 0 48 52
100
Willingness to purchase 0 72 28

Chart Title
80
70
60
50
40
30
20
10
0
Petrol Gas Solar E-rickshaw

The most preferred alternative Willingness to purchase


Predominant
Some what
Not at all

A Little

factor

Total
A lot
Realizing a solar electric rickshaw could cut 2 5 11 39 43 100
pollution significantly
Realizing a solar electric rickshaw could save 0 6 16 34 44
me significantly in monthly fuel costs
Getting a tax benefit and subsidy for 0 8 6 22 64
purchasing a solar e-rickshaw.
Having a 10-year/ 250,000 km solar electric 1 10 29 40 20
rickshaw battery warranty.
Having recharging facilities at home, so that I 0 13 32 36 19
could recharge easily overnight.
Owning a solar electric rickshaw would make 4 14 34 30 18
a statement regarding my strong
environmental values.
Owning a solar electric rickshaw would make 7 21 39 18 15
it clear to others that I am on the forefront of
new technology
Some what
Not at all

Prevent
A Little

Buying

Total
A lot

The lifetime of the battery. 2 7 21 69 11 100

The replacement cost of the battery. 1 10 33 45 11

The potential inconvenience of recharging 1 3 14 47 35


the battery.
Low pick up and power of solar e-rickshaw 2 1 22 45 30

The travelling range of solar electric rickshaw 3 8 57 21 11


after full charging
Loading capacity of solar electric rickshaw 5 18 50 19 8
Difficulties in getting the Solar Electric 6 21 49 14 10
Rickshaw serviced or repaired.

A fully charged electric rickshaw can drive a maximum number of miles


before it has to be charged again. What is the least number of miles that
it has to be able to drive on one battery for you to consider buying an
electric rickshaw?

Frequency Percent Valid Percent Cumulative


Percent

100 1 1.0 1.0 1.0

130 1 1.0 1.0 2.0

150 29 29.0 29.0 31.0

160 3 3.0 3.0 34.0

170 1 1.0 1.0 35.0

Valid 180 11 11.0 11.0 46.0

200 37 37.0 37.0 83.0

250 4 4.0 4.0 87.0

300 12 12.0 12.0 99.0

350 1 1.0 1.0 100.0

Total 100 100.0 100.0

Would you be interested in buying a solar electric rickshaw within the next 2 years?

Frequency Percent Valid Percent Cumulative


Percent

not at all intrested 24 24.0 24.0 24.0

2 21 21.0 21.0 45.0

3 8 8.0 8.0 53.0


4 5 5.0 5.0 58.0

5 7 7.0 7.0 65.0

Valid 6 2 2.0 2.0 67.0

7 8 8.0 8.0 75.0

8 5 5.0 5.0 80.0

9 4 4.0 4.0 84.0

Very intrested 16 16.0 16.0 100.0


Total 100 100.0 100.0
Analysis Case Processing Summary

Unweighted Cases N Percent

Valid 85 85.0
Missing or out-of-range 0 .0
group codes
At least one missing 15 15.0
discriminating variable
Excluded Both missing or out-of-range 0 .0
group codes and at least
one missing discriminating
variable
Total 15 15.0
Total 100 100.0

Group Statistics

Most preffred option Mean Std. Valid N (listwise)


Deviation Unweighted Weighted

SER is more reliable than current rickshaw 2.36 1.224 39 39.000

SER is accelerating slower than current rickshaw 1.92 1.010 39 39.000

SER does less pollution than current rickshaw 1.28 .560 39 39.000

SER have unique features that stand out from 2.21 .833 39 39.000
gasoline counterparts
Gas SER are much cheaper to own in the long run 1.72 1.099 39 39.000
than gasoline rickshaws

SER doesn’t offer enough performance and 2.03 .986 39 39.000


pickup

Gas prices are not high enough to justify buying a 2.67 .955 39 39.000
solar electric rickshaw
SER is more reliable than current rickshaw 1.85 .894 46 46.000
SER is accelerating slower than current rickshaw 2.35 1.140 46 46.000
SER does less pollution than current rickshaw 1.28 .720 46 46.000
SER have unique features that stand out from 2.07 .854 46 46.000
gasoline counterparts
Solar E
SER are much cheaper to own in the long run 1.50 .837 46 46.000
rickshaw
than gasoline rickshaws
SER doesn’t offer enough performance and 2.11 .948 46 46.000
pickup
Gas prices are not high enough to justify buying a 2.50 .863 46 46.000
solar electric rickshaw
Total SER is more reliable than current rickshaw 2.08 1.082 85 85.000
SER is accelerating slower than current rickshaw 2.15 1.097 85 85.000

SER does less pollution than current rickshaw 1.28 .648 85 85.000

SER have unique features that stand out from 2.13 .842 85 85.000
gasoline counterparts

SER are much cheaper to own in the long run 1.60 .966 85 85.000
than gasoline rickshaws

SER doesn’t offer enough performance and 2.07 .961 85 85.000


pickup

Gas prices are not high enough to justify buying a 2.58 .905 85 85.000
solar electric rickshaw

Tests of Equality of Group Means

Wilks' F df1 df2 Sig.


Lambda

SER is more reliable than current rickshaw .944 4.926 1 83 .029


SER is accelerating slower than current rickshaw .962 3.251 1 83 .075
SER does less pollution than current rickshaw 1.000 .000 1 83 .997
SER have unique features that stand out from gasoline .993 .580 1 83 .449
counterparts
SER are much cheaper to own in the long run than gasoline .987 1.075 1 83 .303
rickshaws
SER doesn’t offer enough performance and pickup .998 .156 1 83 .694
Gas prices are not high enough to justify buying a solar electric .991 .714 1 83 .401
rickshaw

In the table ‘Tests of Equality of Group Means’ the results of univariate


ANOVA’s, carried out for each independent variable, are presented. Here, only
“SER is more reliable than current rickshaw” (variable motive) differ (Sig. =
0.029) for the two groups (Gas and Solar E-rickshaw).

Box's Test of Equality of Covariance Matrices

Log Determinants

Most preffred option Rank Log


Determinant

Gas 7 -3.114
Solar E rickshaw 7 -3.310
Pooled within-groups 7 -2.340
The ranks and natural logarithms of determinants
printed are those of the group covariance matrices.

Test Results

Box's M 73.041
Approx. 2.371

df1 28
F
df2 22660.861

Sig. .000

Tests null hypothesis of equal


population covariance matrices.

The significance value of 0.000 indicates that the data is differ significantly
from multivariate normal.

Summary of Canonical Discriminant Functions

Eigenvalues

Function Eigenvalue % of Variance Cumulative % Canonical


Correlation

1 .115a 100.0 100.0 .321

a. First 1 canonical discriminant functions were used in the analysis.

The maximum possible number of discriminant functions is the smaller of one


less than the number of groups defined by the dependent variable and the
number of independent variables. In this analysis there were 2 groups defined
by preferred option and 7 independent variables, so the maximum possible
number of discriminant functions was 1.

For each discriminant function, the Eigenvalue is the ratio of between-group to


within group sums of squares. Large Eigenvalues imply superior functions.

An eigenvalue provides information on the proportion of variance explained. A


canonical correlation of 0.321 suggests the model explains 10.30% (i.e.0.3212
×100) of the variation in the grouping variable.
An eigenvalue indicates the proportion of variance explained. (Between-groups
sums of squares divided by within-groups sums of squares). A large eigenvalue
is associated with a strong function. The canonical relation is a correlation
between the discriminant scores and the levels of the dependent variable. A
high correlation indicates a function that discriminates well. The present
correlation of 0.321 is low (1.00 is perfect).

Wilks' Lambda

Test of Function(s) Wilks' Lambda Chi-square df Sig.

1 .897 8.637 7 .280

H0 : There is no significant discriminating power in the variables.


H1 : There may be a significant discriminating power in the variables.

Here p value = 0.280


So reject the alternative hypothesis H1 and accept the null hypothesis, so there
is no significant discriminating power in variable
Wilks‘ λ for each predictor is the ratio of the within group sum of squares to the
total sum of squares. Its value varies between 0 and 1.
Large values of λ (near 1) indicate that group means do not seem to be different.
Small values of λ (near 0) indicate that the group means seem to be different.
So here value of wilks’ lambda is 0.891 which is higher value so it can be
interpreted that the group means do not seems to be different.
The significance of the discriminant function is indicated by Wilks’ lambda and
provides the proportion of total variability not explained, i.e. it is the converse of
the squared canonical correlation.

A lambda of 1.00 occurs when observed group means are equal (all the
variance is explained by factors other than difference between those means),
while a small lambda occurs when within-groups variability is small compared
to the total variability. A small lambda indicates that group means appear to
differ. The associated significance value indicates whether the difference is
significant. Here, the Lambda of 0.897 has a significant value (Sig. = 0.280);
thus, the group means not appear to differ.
In the table of Wilks' Lambda which tested functions for statistical significance,
the direct analysis identified 1 discriminant functions that were statistically
significant. The Wilks' lambda statistic for the test of function 1 (chi-
square=8.367) had a probability of 0.280 which was more than the level of
significance of 0.05. The significance of the maximum possible number of
discriminant functions doesn’t supports the interpretation of a solution using 1
discriminant function.

Standardized Canonical Discriminant Function Coefficients

Function

SER is more reliable than current rickshaw .572


SER is accelerating slower than current rickshaw -.593
SER does less pollution than current rickshaw -.294
SER have unique features that stand out from gasoline counterparts .247
SER are much cheaper to own in the long run than gasoline rickshaws .415
SER doesn’t offer enough performance and pickup -.242
Gas prices are not high enough to justify buying a solar electric rickshaw .039

Finding the Better Predictors out of the group of 5 Predictor Variables


There are 7 independent (predictor) variables. Now how to decide which of
these is a better predictor of a preferred option of gas or solar e rickshaw. To
answer this, look at the standardized coefficients in Table of the output

This output shows that SER is accelerating slower than my current rickshaw is
the best predictor, with coefficient of -0.593 followed by SER is more reliable
than current rickshaw of 0.572. The absolute value of the standardized
coefficient of each variable indicates its relative importance.
Structure Matrix

Function

SER is more reliable than current rickshaw .719


SER is accelerating slower than current rickshaw -.584
SER are much cheaper to own in the long run than gasoline rickshaws .336
Gas prices are not high enough to justify buying a solar electric rickshaw .274
SER have unique features that stand out from gasoline counterparts .247
SER doesn’t offer enough performance and pickup -.128
SER does less pollution than current rickshaw -.001

Pooled within-groups correlations between discriminating variables and standardized


canonical discriminant functions
Variables ordered by absolute size of correlation within function.

Canonical Discriminant Function Coefficients

Function

SER is more reliable than current rickshaw .540


SER is accelerating slower than current rickshaw -.548
SER does less pollution than current rickshaw -.451
SER have unique features that stand out from gasoline counterparts .293
SER are much cheaper to own in the long run than gasoline rickshaws .429
SER doesn’t offer enough performance and pickup -.250
Gas prices are not high enough to justify buying a solar electric rickshaw .043
(Constant) -.270

Unstandardized coefficients

The ‘Canonical Discriminant Function Coefficients’ indicate the unstandardized


scores concerning the independent variables. It is the list of coefficients of the
unstandardized discriminant equation. Each subject’s discriminant score would
be computed by entering his or her variable values (raw data) for each of the
variables in the equation.

Functions at Group Centroids

Most preferred option Function


1
Gas .364
Solar E rickshaw -.308
Unstandardized canonical
discriminant functions evaluated at
group means

Classification Statistics

Classification Processing Summary

Processed 100
Missing or out-of-range 0
group codes
Excluded
At least one missing 15
discriminating variable
Used in Output 85

Classification Resultsa

Predicted Group Membership Total


Most preferred option Gas Solar E
rickshaw

Gas 24 15 39
Count
Solar E rickshaw 18 28 46
Original
Gas 61.5 38.5 100.0
%
Solar E rickshaw 39.1 60.9 100.0

a. 61.2% of original grouped cases correctly classified.

Table indicates that the discriminant function we have obtained is able to


classify 61.2% of the 85 cases correctly. This figure is in the ‘% correct’ column
of the classification matrix. More specifically, it also says that out of 39 cases
predicted to choose most preferred option gas, 24 were observed to choose
most preferred option as gas. Similarly, from the column 2, we understand that
out of 46 cases predicted to choose solar e rickshaw as a preferred option, only
28 was found to choose solar e rickshaw as most preferred option. Thus, on a
whole 33 case out of 85 was mis-classified by the discriminant model, this gives
us a classification (or predictable) accuracy level of (85-33)/85 or 61.2%. This
level of accuracy may not hold for all future classifications of new cases. But it
is still a pointer towards the model being a good one, assuming the input data
was relevant and scientifically collected.

Discriminant analysis for willingness to purchase

Analysis Case Processing Summary

Unweighted Cases N Percent

Valid 85 85.0
Missing or out-of-range 0 .0
group codes
At least one missing 15 15.0
discriminating variable
Excluded Both missing or out-of-range 0 .0
group codes and at least
one missing discriminating
variable
Total 15 15.0
Total 100 100.0

Group Statistics

Willingness to Purchase Mean Std. Valid N (listwise)


Deviation Unweighted Weighted

SER is more reliable than current rickshaw 2.25 1.129 60 60.000

SER is accelerating slower than current rickshaw 2.00 .939 60 60.000

SER does less pollution than current rickshaw 1.30 .696 60 60.000

SER have unique features that stand out from 2.15 .755 60 60.000
gasoline counterparts
Gas SER are much cheaper to own in the long run 1.60 1.012 60 60.000
than gasoline rickshaws

SER doesn’t offer enough performance and 1.97 .843 60 60.000


pickup

Gas prices are not high enough to justify buying a 2.65 .899 60 60.000
solar electric rickshaw
Solar E SER is more reliable than current rickshaw 1.68 .852 25 25.000
rickshaw SER is accelerating slower than current rickshaw 2.52 1.358 25 25.000
SER does less pollution than current rickshaw 1.24 .523 25 25.000
SER have unique features that stand out from 2.08 1.038 25 25.000
gasoline counterparts
SER are much cheaper to own in the long run 1.60 .866 25 25.000
than gasoline rickshaws
SER doesn’t offer enough performance and 2.32 1.180 25 25.000
pickup
Gas prices are not high enough to justify buying a 2.40 .913 25 25.000
solar electric rickshaw
SER is more reliable than current rickshaw 2.08 1.082 85 85.000

SER is accelerating slower than current rickshaw 2.15 1.097 85 85.000

SER does less pollution than current rickshaw 1.28 .648 85 85.000

SER have unique features that stand out from 2.13 .842 85 85.000
gasoline counterparts
Total SER are much cheaper to own in the long run 1.60 .966 85 85.000
than gasoline rickshaws

SER doesn’t offer enough performance and 2.07 .961 85 85.000


pickup

Gas prices are not high enough to justify buying a 2.58 .905 85 85.000
solar electric rickshaw

Tests of Equality of Group Means

Wilks' F df1 df2 Sig.


Lambda

SER is more reliable than current rickshaw .942 5.134 1 83 .026


SER is accelerating slower than current rickshaw .953 4.115 1 83 .046
SER does less pollution than current rickshaw .998 .150 1 83 .700
SER have unique features that stand out from .999 .121 1 83 .729
gasoline counterparts
SER are much cheaper to own in the long run than 1.000 .000 1 83 1.000
gasoline rickshaws
SER doesn’t offer enough performance and pickup .972 2.426 1 83 .123
Gas prices are not high enough to justify buying a .984 1.353 1 83 .248
solar electric rickshaw
Box's Test of Equality of Covariance Matrices

Log Determinants

Willingness to Purchase Rank Log


Determinant

Gas 7 -3.223
Solar E rickshaw 7 -2.926
Pooled within-groups 7 -2.371

The ranks and natural logarithms of determinants printed


are those of the group covariance matrices.

Test Results

Box's M 63.533
Approx. 2.001

df1 28
F
df2 7897.543

Sig. .001

Tests null hypothesis of equal


population covariance matrices.

Summary of Canonical Discriminant Functions

Eigenvalues

Function Eigenvalue % of Variance Cumulative % Canonical


Correlation

1 .151a 100.0 100.0 .362

a. First 1 canonical discriminant functions were used in the analysis.

The Eigen values are related to the canonical correlations and describe how best
discriminating ability, the functions possess. The % of variances is the discriminating
ability of the 2 groups. Since there is only one function, 100% of the variance is
accounted by this function. The cumulative % of the variance gives the current and
proceeding cumulative total of the variance. As mentioned above, as there is only one
function in the present research we have 100% of the cumulative variance.
The Eigen value gives the proportion of variance explained. A larger Eigenvalue
explains a strong function. The canonical relation is a correlation between the
discriminant scores and the levels of these dependent variables. The higher the
correlations value, the better the function that discriminates the values. 1 is considered
as perfect. Here, we have the correlation of 0.362 is comparatively low.

An eigenvalue provides information on the proportion of variance explained. A


canonical correlation of 0.362 suggests the model explains 13.10% (i.e.0.3622 ×100) of
the variation in the grouping variable.

H0 : There is no significant discriminating power in the variables.


H1 : There may be a significant discriminating power in the variables.

Wilks' Lambda

Test of Function(s) Wilks' Lambda Chi-square df Sig.

1 .869 11.154 7 .132

p value (Sig value of the above output) = .132


Rule: accept Null Hypothesis H0 and reject Alternate Hypothesis H1 Here 0.000 <
0.05, therefore we accept null Hypothesis H0 and reject alternate Hypothesis H1 and
conclude that based on the sample data, there is not statistically significant
discriminating power in the variables included in the model.

The Wilks Lambda is one of the multivariate statistics calculated by SPSS. The lower
the value of Wilks' Lambda, the better. In the present case the value is 0.869. The
Chi-square is 11.154 with 7 degrees of freedom, which is based on the groups present
in the categorical variables. A Wilks Lambda of 1.00 is when the observed group
means are equal, while a small Wilks Lambda is small when the within-groups
variability is small compared to the total variability. This indicates that the group
means appear to differ. Here higher Wilks' Lambda indicate that the group means is
not appear to differ.
Standardized Canonical Discriminant Function Coefficients

Functio
n

SER is more reliable than current rickshaw .690


SER is accelerating slower than current rickshaw -.332
SER does less pollution than current rickshaw -.050
SER have unique features that stand out from gasoline counterparts .103
SER are much cheaper to own in the long run than gasoline rickshaws -.173
SER doesn’t offer enough performance and pickup -.511
Gas prices are not high enough to justify buying a solar electric rickshaw .418

The standardized canonical discriminant function coefficient is used to calculate the


discriminant score. The score is calculated as a predicted value from the linear
regression using the above standardized coefficients and the standarised variables.

This output shows that SER doesn’t offer enough performance and pickup is the best
predictor, with coefficient of -0.511 followed by SER is more reliable than current
rickshaw of 0.6902. The absolute value of the standardized coefficient of each
variable indicates its relative importance

Structure Matrix

Function

SER is more reliable than current rickshaw .641


SER is accelerating slower than current rickshaw -.574
SER doesn’t offer enough performance and pickup -.441
Gas prices are not high enough to justify buying a solar electric rickshaw .329
SER does less pollution than current rickshaw .110
SER have unique features that stand out from gasoline counterparts .098
SER are much cheaper to own in the long run than gasoline rickshaws .000

Pooled within-groups correlations between discriminating variables and standardized


canonical discriminant functions
Variables ordered by absolute size of correlation within function.

Functions at Group Centroids

Willingness to Purchase Function

1
Gas .248
Solar E rickshaw -.594

Unstandardized canonical discriminant


functions evaluated at group means

Classification Statistics

Classification Processing Summary

Processed 100
Missing or out-of-range 0
group codes
Excluded
At least one missing 15
discriminating variable
Used in Output 85

Prior Probabilities for Groups

Willingness to Purchase Prior Cases Used in Analysis

Unweighted Weighted

Gas .500 60 60.000


Solar E rickshaw .500 25 25.000
Total 1.000 85 85.000
It has been observed that 69.4% of data was correctly classified as willingness
to purchase gas and solar e rickshaw by the discriminant function. It has also
been noticed that out of the 60 responses, 42 have been correctly classified as
willingness to purchase gas option. And 18 responses are wrongly classified as
willingness to purchase solar e rickshaw. Out of the 25-solar e rickshaw option,
17 responses have been correctly classified as willingness to purchase solar e
rickshaw whereas 8 responses have been wrongly classified as willingness to
purchase gas option.