Understanding consumer online shopping behaviour from the perspective of transaction costs

Table 5.13 Preliminary Data Analysis- Customer Loyalty (LOY) 
Step One - Exploratory 
Wording EFA 
 
Loadings 
LOY1 
LOY2
LOY3
LOY4 
LOY5 
This online store is my first choice
.88 
LOY1 
1.00 
Recommend this online store
.91 
LOY2 
.820 
1.00 
Encourage other friends 
.91 
LOY3 
.753 
.822 
1.00 
I doubt that I would switch 
.89 
LOY4 
.686 
.718 
.772 
1.00 
This online store is the best one 
.84 
LOY5 
.636 
.653 
.690 
.773 
1.00 
Reliability 
0.93 
KMO 
.867 
Variance Explained 
78.63 
Bartlett’s 
.000 
Step Two – Confirmatory 
Indicators CFA 
 
Stand. Loadings 
Stand. 
Error 
Critical 
Ratio
AVE CR 
LOY1 .87 
-- 
-- 
.70 
.920 
LOY2 .94 
.026 
41.23
*
LOY3 .87 
.027 
36.72
*
LOY4 .77 
.034 
25.55
*
LOY5 .70 
.031 
29.37
*
Model Fit 
Statistic 
Model Fit 
Statistic 
χ
2
8.453 
TLI 
.992 
df
2 
CFI 
.998 
P .015 
RMSEA 
.058 
GFI .997 
RMR 
.019 
* 
p 
< .001

---

241
5.5.13 Preliminary Analysis – Consumer’s Risk-Bearing Propensity 
The consumer’s risk-bearing propensity construct was measure by five items (RISK1 to 
RISK5). The two-step preliminary data analysis process shown in Figure 5.1 was employed 
as a screening mechanism for data measured by these items (refer Table 5.14 for results). 
Evaluation of the correlation matrix through the KMO and Bartlett’s Test results in a high 
KMO statistic (.812) and a significant probability level (p< .001) for the Bartlett’s test. These 
results indicate that sufficient correlations were found within the correlation matrix for factor 
analysis to proceed. In addition, bivariate correlations were inspected and all coefficients fell 
within the acceptable range for factor analysis of .30 and .90. EFA was then conducted which 
produced a single factor structure with strong item loading ranging from .71 to .87 and the 
variance explained was 62.28%. Cronbach’s alpha of .85 indicated good reliability of the 
scale. At this point, as all the items met the criteria of Step One, they were retained for CFA 
analysis in Step Two. 
The five items (retained from Step One) measuring consumer’s risk-bearing propensity were 
subjected to CFA in Step Two (refer Figure 5.1) of the preliminary analysis. The results 
indicated a good fit of the measurement model, for example, 
χ
2
/df (3.867) was less than 5, 
GFI
(.997
), TLI (.986) and CFI (.997) were greater than .90, RMSEA (.055) was less than .08 
and RMR (.020) was less than .05. Moreover, Table 5.14 shows that the standardized factor 
loadings of the five items ranged from .55 to .87. All exceeded the minimum standardized 
loading of .50, with two of them exceeding the ideal criterion of .70 (Hair
 et al.
2006). The 
AVE was .51, exceeding the recommended level of .50 (Hair
 et al.
2006). The CR was .833, 
exceeding the recommended criterion of .70 (Nunnally 1978). As a result, items RISK1 to 
RISK5 were considered to provide a reliable and valid measurement scale for the construct, 

---

242
consumer’s risk-bearing propensity. All five items were retained for the overall measurement 
model.