CHAPTER 4. DATA ANALYSIS AND RESULT
Tải bản đầy đủ
Self-employed (including freelance)
13
4.3%
Part-timer/ Contract/ Temporary employee
13
4.3%
Not yet have a job
Student
13
10
4.3%
3.3%
3
1.0%
1
0.3%
3 - 6 mil. VND
78
26.0%
6 - 10 mil. VND
77
25.7%
10 - 15 mil. VND
15 - 20 mil. VND
72
21
24.0%
7.0%
20 - 25 mil. VND
9
3.0%
25 - 30 mil. VND
30 - 35 mil. VND
21
5
7.0%
1.7%
35 - 40 mil. VND
5
1.7%
40 - 45 mil. VND
1
0.3%
10
3.3%
2 members
28
9%
3 members
4 members
30
64
10%
21%
5 members
83
28%
6 members
48
16%
7 members
8 members
30
12
10%
4%
9 members
2
1%
10 members
3
1%
Housewife
Household monthly income
Less than 3 mil. VND
50 mil VND or more
Number of members in household
As can be seen, There were 143 males (48%) and 157 females (53%) participating the
study. Most of qualified respondents (54%) belonged to 26-34 years old, 44% were 1825 years old. Only 2% were more than 35 years old.
All respondents could recall their most-often-purchase retailer. Nguyen Kim, Thien Hoa
and Cho Lon were the three most popular merchandisers to be recalled with 41%, 18%
and 13%, respectively. There were about 8% of respondent who purchase most
frequently at the “other” small retailers.
Regarding demographic information, there were 46 respondents (15%) married. The
result also recorded 206 respondents (69%) to be employees of private or foreign
43
company and 42 people (14%) working for government enterprise and 9% unemployed
cases (yet have a job, student or housewife).
In average, there were about five members (4.86) in an urban household. 13.7 million
VND was the average household monthly income. This number reflected the living
standard of household in urban of Ho Chi Minh only and could not be projected to
Vietnam nationwide
Table 4-2: Average number in one household and average household monthly income
N
Minimum
Maximum
Mean
Number of members in household
300
2 members
10 members
4.86 members
Average household monthly income
300
2,900,000
70,000,000
13,699,667
4.2 Cronbach alpha reliability analysis
Table 4-3 represented the results of Cronbach alpha analysis. As can be seen, the general
alphas range from 0.7943 (store convenience) to 0.9636 (store service). Also, the general
alpha of each independent variable was higher than alpha-if-item-deleted of every item
in the construct. In summary, all of the construct used in the main survey were reliable in
an acceptable range of Cronbach alpha.
Table 4-3: Cronbach alpha result for main survey
Scale
Mean if
Item
Deleted
Scale
Variance
if Item
Deleted
Corrected
Item-Total
Correlation
Alpha if
Item
Deleted
Store Service: Alpha = 0.9636
EMPL1_The store's employees are very friendly
24.3967
46.8488
0.8286
0.9618
EMPL2_The store's employees are warm-hearted in
consulting customers
24.3467
45.6052
0.9091
0.9529
EMPL3_The store's employees are honest with
customers
24.4600
46.8178
0.8420
0.9602
24.2700
46.5121
0.8721
0.9569
24.1000
46.6120
0.9326
0.9508
23.9600
46.7810
0.9131
0.9528
41.6871
0.8654
0.9073
AFTE1_Free of on-site delivery and assembling
AFTE2_Offer considerate warranty service
AFTE3_After-sales service staffs are friendly and
helpful
Store Facilities: Alpha = 0.9301
FACI1_Facilities for rest at the store are
comfortable
23.9567
44
FACI2_Elevator, rest room and lightings at the store
are comfortable
FACI3_The placement of the aisles is appropriate
ATMO1_Has eye-catching interior design
ATMO2_Products display is appropriate
ATMO3_In-door movement is easy
24.2200
41.4297
0.7803
0.9187
24.3500
41.2383
0.8330
0.9112
24.5700
43.0887
0.7370
0.9239
24.2667
44.1159
0.7760
0.9190
24.3033
42.8274
0.7798
0.9182
Product and Marketing Communications: Alpha = 0.8988
MERC1_The store has wide range of brand and
product
29.3767
51.1185
0.6757
0.8852
MERC2_The store usually has products of new
arrival
29.4167
50.4245
0.7574
0.8763
MERC3_The store always has product I need (no
out-of-stock)
30.0800
52.1207
0.5856
0.8959
MERC4_The store offer prestige and high quality
products
29.7200
50.0150
0.7665
0.8751
PROM1_The store has attractive promotions
programs
29.5700
49.3295
0.7715
0.8741
PROM2_The store often celebrate events for loyal
customers
30.0833
50.1034
0.6713
0.8861
29.8533
49.5503
0.6937
0.8834
8.6900
9.1711
0.5633
0.8205
9.2633
6.7164
0.7344
0.7095
8.9933
6.8227
0.7752
0.6640
9.8733
8.7464
0.5691
0.7893
CONV2_The store has convenient parking lot and
helmet keeping
10.2900
7.5109
0.6863
0.6646
CONV3_The opening hours of the store is
convenient
10.1167
8.4245
0.6599
0.6971
8.8367
5.3545
0.6383
0.7662
9.0500
5.5661
0.7829
0.6156
PROM3_The store has many advertising programs
Perceived Price: Alpha = 0.8249
PRIC1_The store offer value-for-money products
PRIC2_I can buy products at cheaper price at the
store
PRIC3_The prices at the store are very competitive
Store Convenience: Alpha = 0.7943
CONV1_The store is easily accessible
Consumer loyalty: Alpha = 0.8131
LOYA1_The store would be my first choice
LOYA2_I consider myself loyal to the store
45
LOYA3_Even in the case of price increase, I also
would like to buy products from the store
9.4000
6.3478
0.5713
0.8128
4.3 Exploratory factor analysis
The following rotated component matrix (Table 4-4) showed the results of exploratory
factor analysis with principle component extraction and varimax rotation method.
As indicated, all of items were loaded appropriately in their construct from 0.44
minimum score (The store always has product I need) to 0.88 greatest score (Offer
considerate warranty service). Although the loading score of MERC_3 (The store always
has product I need) was less than 0.5. This item was not too small to be removed out of
the study because of its content value in merchandising factor (Nguyen, 2011)
46
Table 4-4: EFA rotated component matrix for main survey
Item
1
AFTE2_Offer considerate warranty service
0.88
EMPL2_The store's employees are warm-hearted in consulting customers
0.84
AFTE3_After-sales service staffs are friendly and helpful
0.84
EMPL3_The store's employees are honest with customers
0.82
AFTE1_Free of on-site delivery and assembling
0.80
EMPL1_The store's employees are very friendly
0.76
Component
2
3
4
FACI1_Facilities for rest at the store are comfortable
0.79
FACI3_The placement of the aisles is appropriate
0.77
FACI2_Elevator, rest room and lightings at the store are comfortable
0.75
ATMO1_Has eye-catching interior design
0.74
ATMO3_In-door movement is easy
0.70
ATMO2_Products display is appropriate
0.68
PROM1_The store has attractive promotions programs
0.79
PROM2_The store often celebrate events for loyal customers
0.73
PROM3_The store has many advertising programs
0.72
MERC2_The store usually has products of new arrival
0.71
MERC4_The store offer prestige and high quality products
0.58
MERC1_The store has wide range of brand and product
0.55
MERC3_The store always has product I need (no out-of-stock)
0.44
CONV2_The store has convenient parking lot
0.79
CONV3_The opening hours of the store is convenient
0.75
CONV1_The store is easily accessible
0.69
5
PRIC2_I can buy products at cheaper price at the store
0.85
PRIC3_The prices at the store are very competitive
0.83
PRIC1_The store offer value-for-money products
0.51
47
Table 4-5: Total variance explained
Initial Eigen Values
Component
Total
% of
Cumulative
Variance
%
Extraction Sums of Squared
Loadings
% of
Cumulative
Total
Variance
%
Rotation Sums of Squared
Loadings
% of
Cumulative
Total
Variance
%
1
12.59
50.35
50.35
12.59
50.35
50.35
5.29
21.16
21.16
2
2.04
8.17
58.52
2.04
8.17
58.52
4.28
17.12
38.28
3
1.68
6.72
65.24
1.68
6.72
65.24
3.93
15.74
54.01
4
1.36
5.42
70.66
1.36
5.42
70.66
2.76
11.04
65.06
5
1.16
4.64
75.31
1.16
4.64
75.31
2.56
10.25
75.31
6
0.82
3.30
78.60
7
0.67
2.70
81.30
8
0.65
2.58
83.88
9
0.48
1.92
85.81
10
0.44
1.76
87.57
11
0.41
1.62
89.19
12
0.34
1.38
90.57
13
0.31
1.24
91.81
14
0.29
1.15
92.96
15
0.27
1.09
94.05
16
0.26
1.05
95.11
17
0.24
0.97
96.08
18
0.21
0.83
96.91
19
0.19
0.76
97.67
20
0.18
0.72
98.39
21
0.14
0.55
98.94
22
0.13
0.53
99.47
23
0.06
0.23
99.70
24
0.04
0.18
99.88
25
0.03
0.12
100.00
48
Table 4-5 represented Total Variance Explained from the EFA. The Total Variance
Explained of five factors was 75.31%, higher than 60%, a standard of a good analysis
(Nguyen, 2011). This meant that the five explored factors explained 75.31% variance of
the observed variables.
Extraction Method: Principal Component Analysis.
Table 4-6: KMO and Bartlett’s Test
Kaiser-Meyer-Olkin Measure of
Sampling Adequacy.
0.9065
Approx. ChiSquare
Bartlett's Test of Sphericity
7,308
Df
300
0.000
Sig.
Table 4-6 showed the results of KMO and Bartlett test. KMO was greater than 90%,
indicating a very-good EFA result (Kaiser, 1974). Also, P value of Bartlett test was less
than 5%, concluding that the correlation matrix among variables was not an identity
matrix and indicating the correlation among variables.
4.4 Multiple linear regression
Table 4-7: Regression model summary
R
0.822
Change Statistics
R
Square
Adjusted
R
Square
Std. Error of
the Estimate
0.676
0.670
0.657
R
Square
Change
0.676
F
Change
df1
122.495
5.000
df2
294.000
Sig. F
Change
0.000
a. Predictors: (Constant), Perceived Price, Store Convenience, Store Facilities, Store Service,
Product and Marketing Communications
b. Dependent Variable: Consumer loyalty
Table 4-7 showed the regression model summary in which R square and adjusted R
square were represented. R square equal to 0.676, indicating that 68% variability of
consumer loyalty was predicted from five dependent variables. Adjusted R square
(0.670) was really closed to R square; the difference was 0.676 – 0.670 = 0.6%. This
meaning that if the model is derived from the population rather than a sample, it would
account for 0.6% less variance in the outcome of dependent variable.
49
Table 4-8: ANOVA F test
Sum of
Squares
Df
Mean Square
Regression
264.496 5
52.899
Residual
126.963 294
0.432
Total
391.459 299
F
Sig.
0.000a
122.495
a. Predictors: (Constant), Perceived Price, Store Convenience, Store Facilities, Store Service,
Product and Marketing Communications
b. Dependent Variable: Consumer loyalty
Another output of linear regression was analysis of variance (ANOVA) which can be
seen in the table 4-8. This analysis tests whether the model is significantly good at
predicting the outcome. F-ratio represents the ratio of the improvement in predicting the
result from fitting model (regression), to the inaccuracy that still exists in the model
(residual). F value was much greater than 1, indicating that the improvement from fitting
the regression model was much higher than the inaccuracy within the model. P value
was less than 0.001, thus we can conclude that the model is significantly better at
predicting the consumer loyalty outcome.
Table 4-9: Multiple linear regression result for main survey
Model
Unstandardized
Coefficients
T
0.248
0.143
0.285
Std.
Error
0.184
0.041
0.046
0.170
0.322
1.345
3.456
6.165
0.201
0.052
0.206
0.106
0.159
0.035
0.037
0.127
0.183
B
(Constant)
Store Service
Store Facilities
Product and
Marketing
Communications
Store Convenience
Perceived Price
Standardized
Coefficients
Dependent Variable: Consumer loyalty
Beta
Collinearity
Statistics
Sig.
Tolerance
VIF
0.180
0.001
0.000
0.456
0.404
2.192
2.475
3.898
0.000
0.394
2.540
3.064
4.306
0.002
0.000
0.647
0.609
1.546
1.641
Finally, the result of regression was revealed in the above table 4-9. As can be seen,
Store Service, Sore Facilities, Product and Marketing Communications, Store
Convenience and Perceived Price have positive standardized beta and p value < 0.005,
50
indicating a significant positive relationship between each of them and dependent
variable (consumer loyalty)
The result of regression model was also generated. The following figure 4-1 showed that
consumer loyalty (Y) is determined by five independent factors Store Service (X1),
Store Facilities (X2), Product & Marketing Communication (X3), Store Convenience
(X4) and Perceived Price (X5). The relationship between independent and dependent
variables is thus described in the following equation
Y = 0.170X1 + 0.322X2 + 0.206X3 + 0.127X4 + 0.183X5
X1: Store Service
β1= 0.170
Adjusted R2 = 0.670
X2: Store facilities
β2= 0.322
Y: Consumer
Loyalty
β3= 0.206
X3: Product &
marketing
communication
Y = 0.170X1 + 0.322X2 +
X4: Store convenience
β4= 0.127
0.206X3 + 0.127X4 + 0.183X5
X5: Perceived price
β5= 0.183
Figure 4-1: Result of multiple linear regression
4.5 Hypotheses assessment and discussion
Hypothesis H1b: The store service positively affects consumer loyalty
In Table 4-7, Store service had positive standardize beta (0.170) and P value < 0.005.
We can conclude that store service positively affected consumer loyalty. Hypothesis 1b
was thus supported. Jinfeng and Zhilong (2009) also confirmed the significant
relationship between employee service and consumer loyalty. Our study generalized the
51
result to accept the positive relationship of store service including employee service &
after-sales service and consumer loyalty
Hypothesis H2b: There is a positive relationship between store facilities and consumer
loyalty
Store facilities, in the same way, was posited to positively affect consumer loyalty with
standardized beta = 0.322 and P value = 0.001. We consequently accepted the H2b.
While only physical facilities is supported to significantly correlate with consumer
loyalty (Jinfeng and Zhilong, 2009), our research provided further conclusions about the
combined effects of both physical facilities and store atmosphere on consumer loyalty of
a store
Hypothesis H3b: Product & marketing communication is positively related to
consumer loyalty
The result in table 4-7 illustrated for the positive relationship between merchandising &
marketing communication and consumer loyalty (Standardized beta = 0.206, P value
<0.005). H3b was thus supported. Besides, this conclusion also complemented the result
of Thang and Tan study (2003) which identifies the two separate relationship between
merchandising and promotions towards store preference
Hypothesis H4b: The store convenience positively related to consumer loyalty
Store convenience was recorded to have the smallest standardize beta (0.127). However,
this factor was still posited to positively related to consumer loyalty (P value = 0.002).
This indicated that, the location and the accessibilities of a store also play an important
part in shaping consumer loyalty. Jinfeng and Zhilong (2009) also find the relationship
in their recent retailer equity study. We thus accepted the H4b
Hypothesis H5b: There is a positive relationship perceived price and consumer loyalty
Finally, H5b was also supported when perceived price was recorded to have 0.183 beta
score and P value < 0.005. This result explained an undeniable fact in many emerging
countries that price consideration always stays at one the top priorities for consumers to
select a store and purchase products.
52
Generally, the study satisfactorily achieved research objectives by identifying the
relationship between key store images and consumer loyalty. Statistically, the research
fulfill most of requirements about reliability such as
• Cronbach’s alpha represent a good reliability (0.79 – 0.96)
• The multi-collinearity of all five independent variables in regression model was in an
acceptable range (1.546 – 2.540), indicating that the model typically accept certain
degree collinearity between the predictors
• Adjusted R Square is closed to R square, ensuring the reliable estimation from
population
Additionally, the study also found that store images dimensions affect differently on
consumer loyalty. The results of regression revealed the derived importance score of
each factors based on standardized beta. As can be seen, store facilities had the strongest
impact on consumer loyalty with beta score = 0.322. The second most important factor
was product & marketing communication (0.206), followed by perceived price (0.183)
and store service (0.170). Store convenience was indentified to be the least important
factor with 0.127 beta score. Meanwhile, Jinfeng and Zhilong (2009) explored that store
convenience was the strongest factor impacting consumer loyalty, followed by physical
facilities, employee service and perceived price.
4.6 Summary
This chapter represented the results of the main survey including sample descriptive
information, Cronbach alpha reliability analysis, exploratory factor analysis, multiple
linear regression, hypotheses assessment and discussion.
Descriptive information of the sample showed the demographic characteristics such as:
gender, age, most-often-purchased store, marital status, occupation and household
monthly income. Cronbach alpha results revealed that all items were used in an
acceptable reliability (general alpha of each factor ranged from 0.7943 to 0.9636). All
five factors were explored to be appropriately loaded into their predefined constructs.
Finally, multiple linear regression showed the significant relationship of store service,
store facilities, product & marketing communication, store convenience and perceived
53