Understanding consumer online shopping behaviour from the perspective of transaction costs

Table 5.9 Preliminary Data Analysis- Privacy and Security Concerns (PSC) 
Step One - Exploratory 
Wording EFA 
 
Loadings 
PSC1 PSC2 PSC3 
PSC4 PSC5 
Concerned over the security of personal 
information exchange via the Internet 
.85 PSC1 
1.00 
Concerned that my personal information 
may be shared 
.93 PSC2 
.790 
1.00 
Concerned that it may be read or stored by 
some other person/entity 
.92 
PSC3 .702 .853 1.00
Worried about the security of financial 
transactions 
.89 
PSC4 .632 .760 .818 1.00
Uncomfortable giving my credit card 
number via the Internet 
.85 
PSC5 .653 .705 .711 .738 1.00 
Reliability 
0.93 
KMO 
.865 
Variance Explained 
79.00 
Bartlett’s 
.000 
Step Two – Confirmatory 
Indicators CFA 
 
Stand. Loadings 
Stand. 
Error 
Critical 
Ratio
AVE CR 
PSC1 .73 
-- 
-- 
.71 
.922 
PSC2 .89 
.035 
35.81
*
PSC3 .96 
.047 
29.31
*
PSC4 .85 
.046 
26.91
*
PSC5 .74 
.040 
26.32
*
Model Fit 
Statistic 
Model Fit 
Statistic 
χ
2
4.251 
TLI 
.975 
df
3 
CFI 
.971 
P .519 
RMSEA 
.026 
GFI .968 
RMR 
.005 
* 
p 
< .001
5.5.9 Preliminary Analysis – Environmental Uncertainty
 
The environmental uncertainty construct was measured by seven items (EUN1 to EUN7). 
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 5.10 for results). 
Evaluation of the correlation matrix through the KMO and Bartlett’s Test results in a high 
KMO statistic (.856) 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 

---

232
produced a single factor structure with strong item loading ranging from .69 to .86 and the 
variance explained was 62.50%. Cronbach’s alpha of .90 was computed indicating 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 seven items (retained from Step One) measuring environmental uncertainty were 
subjected to CFA in Step Two (refer Figure 5.1) of the preliminary analysis. The model fit 
indices indicated that this measurement model poorly fit the data (
χ
2
/df = 50.468 > 5, GFI 
= .812 < .90, TLI = .757 < .90, CFI = .838 < .90, RMSEA = .227 < .08, and RMR 
=.174 > .05). After the initial model assessment, the model was modified according to the 
modification indices to improve model fit (Hair
 et al.
2006). Indicators which were related to 
larger reductions of chi-square were identified and eliminated one by one. After removing 
three items (EUN5, EUN6, EUN7), the revised model of environmental uncertainty provided 
a better fit to the data than the initial model. For example, 
χ
2
/df (1.729) was less than 5, GFI 
(.998), TLI (.998) and CFI (.997) were above .90, RMSEA (.028) was less than .08 and RMR 
(.009) was less than .05. As displayed in Table 5.10, the standardized factor loadings of the 
remaining four items ranged from .67 to .96. All exceeded the acceptable level of .50 with 
three of them above the preferable cutoff of .70 (Hair
 et al.
2006). The AVE for this construct 
was .67, exceeding the recommended level of .50 (Hair et al., 1995). The CR was .889, 
exceeding the recommended benchmark of .70 (Nunnally 1978). Taken together, items EUN1 
to EUN4 were considered to provide a reliable and valid measurement scale for 
environmental uncertainty. Therefore, these four items were retained for constructing the 
overall measurement model. 

---

233