Understanding consumer online shopping behaviour from the perspective of transaction costs

Table 5.29 Results of the Structural Model – Search Products 
Equation Predicted 
Variables 
Predictor 
Variables 
Hypothesis 
Beta weight 
T value 
R
2
 
1 
Perceived Consumer TCs 
Internet Access Availability 
H1a(s) -.056 
-1.637 
(n.s.) 
.92 
Perceived Internet Expertise 
H1b(s) -.108 -3.091
**
Online Buying Frequency 
H1c(s) -.244 -5.006
***
Product Quality Concern 
H2a(s) -.012 -.202 
(n.s.) 
E- Service Quality 
H2c(s) -.578 -4.825
***
Reputation of Online Store 
H2d(s) -.128 -2.496
*
Perceived Convenience 
H3a(s) -.083 -2.103
*
Privacy and Security Concerns 
H3b(s) .036 .736 
(n.s.) 
Age
.047 
1.228 
(n.s.) 
Gender
-.016 
-.518 
(n.s.) 
Income
-.012 
-.324 
(n.s.) 
Education Level 
-.013 
-.375 (n.s.) 
2 
Online Purchase Behaviour 
Perceived Consumer TCs 
H4a(s) 
-.690 
-13.328
***
.48 
3 
Customer Loyalty 
Perceived Consumer TCs 
H4b(s) 
-.387 
-5.859
***
.55 
Customer Satisfaction 
H5(s) 
.391 
6.461
***
4 
Customer Satisfaction 
Perceived Consumer TCs 
H4c(s) 
-.800 
-15.148
***
.64 
AVA
.65 
Note:
AVA: Average Variance Accounted for.
*** p < 0.001, ** p < 0.01 , * p < 0.05, ns: p > .05 

---

278
-.690 (-13.328
***
)
-.800 (-15.148
***
)
R
2
= 92%
R
2
= 48%
R
2
= 64% 
R
2
= 55% 
Figure 5.4 Model Results – Search Products 
 
 
 
 
Note: *** p < 0.001, ** p < 0.01, * p < 0.05
-.387 (-5.859
***
)
Consumer-related characteristics 
Internet Access Availability 
Perceived Internet Expertise
Online Buying Frequency
Online store- and product-related 
characteristics 
Product Quality Concern 
E- Service Quality
Reputation of Online Store
Online channel-related 
characterises 
Perceived Convenience 
Privacy and Security Concerns
Perceived 
Consumer TCs of 
Online Shopping
Online 
Purchase 
Behaviour 
Customer 
Satisfaction 
Customer 
Loyalty 
-.108 (-3.091
***
)
-.056 (-1.637)
.036 (.736)
-.012 (-.202)
-.578 (-4.825
***
)
-.128 (-2.496
*
)
-.244 (-5.006
***
)
.391 (6.461
***
)
Age
Gender 
Income 
Education 
.047 (1.288) -.016 (-.518)
-.012 (-.324)
-.013 (-.375)
-.083 (-2.103
*
)

---

279
5.10.2 Model Results --- Experience Products 
The fit indices demonstrated a good fit for this model (experience products), for example, 
χ
2 
of 247.456 with 99 df, p < .01, 
χ
2
/df = 2.500 < 5, GFI = .955 > .90, CFI = .973 > .90, TLI 
= .948 > .90, and RMSEA = .054 < .08. RMR = .046 < .05. Table 5.30 showed the results of 
the SEM analysis (Figure 5.5) for the sample pertaining to experience products by providing 
the path coefficients, average variance accounted (AVA) for, T value and R
2
. All numbered 
hypotheses contain an additional “e” which denotes the use of sample pertaining only to 
experience products in this analysis.
As can be found in Table 5.30, the individual R
2
are greater than the recommended .10 (Falk 
and Miller 1992) for all of the predicted variables and the AVA for the endogenous variables 
was .74. Moreover, all but one path met the criterion of T value (above 1.96), which indicated 
that the eleven paths had significant effects. By examining the standardized path estimates for 
this model results, it was found that all eleven paths were significant in the hypothesized 
direction. Thus, H1a(e), H1b(e), H1c(e), H2a(e), H2c(e), H3a(e), H3b(e), H4a(e), H4b(e), 
H4c(e) and H5(e) were all supported. Nevertheless, in terms of the T value, one path from 
reputation of online store to consumer TCs was not significant in the expected direction since 
the corresponding t value was less than 1.96, and as such H2d(e) was rejected. Furthermore, 
the data showed that 96 per cent of the variance in consumer TCs was explained by seven 
antecedent variables (including internet access availability, perceived internet expertise, 
online buying frequency, product quality concern, e-service quality, perceived convenience, 
and privacy and security concerns), while consumer TCs explained 68 per cent of the 
variance in customer satisfaction. Customer satisfaction and TCs accounted for 73 per cent of 
the variance in customer loyalty. In addition, 60 per cent of the variance in online purchase 
behaviour was explained by TCs.

---

280
The preceding SEM analysis of the proposed model for experience products revealed support 
for all but one hypothesis indicating that the model was supported in the context of 
experience products. Figure 5.5 presents the proposed model for experience products 
showing all path coefficients and R
2
values for endogenous variables.