Understanding consumer online shopping behaviour from the perspective of transaction costs

Table 5.2 Preliminary Data Analysis- Internet Access Availability (IAA) 
Step One - Exploratory 
Wording EFA 
 
Loadings 
IAA1 IAA2 IAA3 IAA4 
Can easily find Internet
.82 
IAA1 
1.00 
PC can easily connect to the Internet 
.88 
IAA2 .676 1.00
Easily access to Internet in study/work place 
.87 
IAA3 
.572 .709 1.00
Access to the Internet at anytime/ 
anywhere .78 
IAA4 .501 .536 .618 1.00 
Reliability 
0.86 
KMO 
.787 
Variance Explained 
70.29 
Bartlett’s 
.000 
Step Two – Confirmatory 
Indicators CFA 
 
Stand. Loadings 
Stand. 
Error 
Critical 
Ratio
AVE CR 
IAA1 .68 
-- 
-- 
.63 
.834 
IAA2 .85 
.031 
40.51
*
IAA3 .84 
.031 
40.51
*
IAA4 n/a 
n/a 
n/a 
Model Fit 
Statistic 
Model Fit 
Statistic 
χ
2
.046 
TLI 
0.992 
df
2 
CFI 
0.983 
P .965 
RMSEA 
.002 
GFI 0.986 RMR 
.002 
* 
p 
< .001
5.5.2 Preliminary Analysis - Perceived Internet Expertise 
The perceived Internet expertise construct was measured using four items (EXP1 to EXP4). 
The two-step preliminary data analysis process shown in Figure 5.1 was employed as a 
screening mechanism for data measured by these item (refer Table 5.3 for results). Evaluation 
of the correlation matrix through the KMO and Bartlett’s Test results in a high KMO statistic 
(.825) 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 .84 to .93 and the 
variance explained was 78.52%. Cronbach’s alpha of .91 was computed indicating good 

---

218
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. 
In Step Two (refer Figure 5.1), CFA was conducted on the four items retained from Step One 
which resulted in all computed statistics indicating a good fit for the measurement model. For 
example, 
χ
2
/df (.066) was less than 5, GFI (0.979), TLI (0.988) and CFI (0.985) were 
above .90, RMSEA (.002) was less than .08 and RMR (.001) was less than .05. Moreover, 
Table 5.3 shows that the standardized factor loadings of the four items ranged from .71 to .90, 
all exceeding the preferable criterion of .70 (Hair
 et al.
2006). The AVE was .70, which 
exceeded the recommended level of .50 (Hair
 et al.
2006). The CR of this construct was .902, 
exceeding the recommended benchmark of .70 (Nunnally 1978). Taken together, items EXP1 
to EXP4 were considered to provide a reliable and valid measurement scale for the construct, 
perceived Internet expertise. Thus, all four items were retained for the overall measurement 
model.