You are on page 1of 8

Analysis Case Processing Summary

Unweighted Cases N Percent

Valid Both
At missing
least one or 145 100
Missing or out-of-
out-of-range group
missing
range group
codes and atcodes
discriminatingleast 0 0
one missing
variable 0 0
discriminating
variable 0 0
Excluded Total 0 0
Total 145 100

Group Statistics
OSA Mean Std. Deviation Valid N (listwise)
Unweighted

AGE 18.632 1.3373 95


GENDER 1.558 0.4993 95
INC 5.463 2.0096 95
EDU 4.168 1.9659 95
PEV 4.789 1.929 95
COM 5.211 1.7436 95
Adopt SHE 4.926 1.7088 95
AGE 18.4 0.7284 50
GENDER 1.42 0.4986 50
INC 6.86 1.7729 50
EDU 6.52 1.4462 50
PEV 6.84 1.5954 50
COM 6.86 1.512 50
Refusal SHE 4.74 2.0683 50
AGE 18.552 1.1663 145
GENDER 1.51 0.5016 145
INC 5.945 2.0371 145
EDU 4.979 2.1196 145
PEV 5.497 2.062 145
COM 5.779 1.8389 145
Total SHE 4.862 1.8357 145

Tests of Equality of Group Means


Wilks' Lambda F df1 df2

AGE 0.991 1.294 1 143


GENDER 0.983 2.501 1 143
INC 0.893 17.128 1 143
EDU 0.72 55.615 1 143
PEV 0.775 41.509 1 143
COM 0.817 32.039 1 143
SHE 0.998 0.336 1 143

Log Determinants
OSA Rank Log Determinant
Adopt 7 4.164
Refusal 7 1.296
Pooled within-
groups 7
The ranks and natural logarithms of determinants 3.674
printed
are those of the group covariance matrices.

Test Results
Box's M 70.448
Approx. 2.365
df1 28
df2 36129.089
F
Tests null hypothesisSig.
of equal population covariance 0
matrices.

Eigenvalues
Function Eigenvalue % of Variance Cumulative % Canonical
Correlation
1 .480a 100 100 0.57
a. First 1 canonical discriminant functions were used in the analysis.

Wilks' Lambda
Test of Function(s) Wilks' Lambda Chi-square df Sig.
1 0.675 54.734 7 0

Standardized Canonical
Discriminant Function Function
Coefficients
1
AGE -0.027
GENDER -0.169
INC 0.074
EDU 0.585
PEV 0.242
COM 0.308
SHE -0.027

Structure Matrix
Function
1
EDU 0.9
PEV 0.777
COM 0.683
INC 0.499
GENDER -0.191
AGE -0.137
SHE -0.07
Variables ordered by absolute size of
correlation within function.

Canonical Discriminant Function


CoefficientsFunction
1 4. Assume that there are two po
AGE -0.024 which group (adopting or refusal) e
GENDER -0.338 From Canonical Discrimination Functio
INC 0.038
OSA = 0.038INC + 0.324EDU + 0.133
EDU 0.324
PEV 0.133 So, for A, we have: OSA = 0.038*2 + 0
COM 0.185 and for B, we have: OSA = 0.038*6 +
SHE -0.015 So, they are both adopting.
(Constant) -2.621
Unstandardized coefficients

Functions at Group Centroids


OSA Function
1
Adopt -0.499
Refusal 0.949
Unstandardized canonical discriminant
functions evaluated at group means

Prior Probabilities for Groups


OSA Prior Cases Used in Analysis
Unweighted Weighted
Adopt 0.5 95 95
Refusal 0.5 50 50
Total 1 145 145

Classification Function Coefficients


OSA
Adopt Refusal
AGE 15.883 15.849
GENDER 14.182 13.692
INC 2.917 2.973
EDU -0.227 0.243
PEV 2.722 2.914
COM -0.1 0.168
SHE 1.35 1.329
(Constant) -176.782 -180.904
Fisher's linear discriminant functions
Valid N (listwise)
Weighted

95
95
95
95
95
95
95
50
50
50
50
50
50
50
145
145
145
145
145
145
145

Sig. 1. Which factors significantly explain for the differences between online
shopping adopting group and online shopping refusing group?
0.257
0.116 Firstly, from Wilk's Lambda, we have the significant at 0 < 0.05, so it's relevent for
0 us to use this data. Based on the tests of equality of group means, we can see
that INC, EDU, PEV and COM has significant at 0 < 0.05, so these factors are
significantly explaining for the differences between online shopping adpotinf
group and online shopping refusing group. Meanwhiles, AGE, GENDER and SHE
have the significant > 0.05, so these factors have no significantly explaining fro
the differences between online shopping adopting group and online shopping
refusing group.
1. Which factors significantly explain for the differences between online
shopping adopting group and online shopping refusing group?

Firstly, from Wilk's Lambda, we have the significant at 0 < 0.05, so it's relevent for
us to use this data. Based on the tests of equality of group means, we can see
that INC, EDU, PEV and COM has significant at 0 < 0.05, so these factors are
0 significantly explaining for the differences between online shopping adpotinf
0 group and online shopping refusing group. Meanwhiles, AGE, GENDER and SHE
have the significant > 0.05, so these factors have no significantly explaining fro
0 the differences between online shopping adopting group and online shopping
0.563 refusing group.

2. How many percentages of the differences between these two groups of


consumers are explained by the predictors?

We have: (Canonical Correlation) ^ 2 = (0.57)^2 = 32.49%


So, 32.49% of the differences between these two groups of consumers are explained by
the predictors.

3. Among the significant factors, which one contributes most and least to the differences
between adopting and refusing group?

Based on the Standardized Canonical Discrimination Function Coefficients, we have EDU has the largest
function at 0.585, so EDU contributes most to the differences between adopting and refusing group.
Meanwhiles, AGE and SHE have the smallest function at -0.027, it indicates that these factors contributes
least to the differences between adopting and refusing group.
Assume that there are two potential consumers having following characteristics, identify
h group (adopting or refusal) each of them may belong to.
Canonical Discrimination Function Coefficients, we have the formulate:

= 0.038INC + 0.324EDU + 0.133PEV + 0.185COM - 2.621

r A, we have: OSA = 0.038*2 + 0.324*5 +0.133*3 + 0.185*8 -2.621 = 0.954


or B, we have: OSA = 0.038*6 + 0.324*7 +0.133*6 + 0.185*3 -2.621 = 1.228
ey are both adopting.
nces between online
g group?

0 < 0.05, so it's relevent for


oup means, we can see
, so these factors are
ine shopping adpotinf
s, AGE, GENDER and SHE
gnificantly explaining fro
up and online shopping
en these two groups of

consumers are explained by

You might also like