Understanding consumer online shopping behaviour from the perspective of transaction costs

Table 5.3 Preliminary Data Analysis- Perceived Internet Expertise (EXP) 
Step One - Exploratory 
Wording 
EFA
Loadings 
 
EXP1 
EXP2 EXP3 EXP4 
Easily use the Web to find product information
.84 EXP1 
1.00 
Get to a specific Website with a browser
.89 EXP2 
.614 
1.00 
Feel comfortable searching the World Wide Web 
.90 
EXP3 
.639 
.776 1.00
Be able to use Web to locate retail sites 
.92 
EXP4 
.737 
.741 .768 1.00 
Reliability 
0.91 
KMO 
.825 
Variance Explained 
78.52 
Bartlett’s 
.000 
Step Two – Confirmatory 
Indicators 
CFA
Stand. Loadings 
Stand. 
Error 
Critical Ratio AVE 
CR 
EXP1 
.71 -- 
-- 
.70 
.902 
EXP2 
.86 .046 
24.78
*
EXP3 
.90 .047 
25.29
*
EXP4 
.86 .038 
30.27
*
Model Fit 
Statistic Model 
Fit 
Statistic 
χ
2
.131 TLI 
0.988 
df
2 CFI 
0.985 
P 
.860 RMSEA .002 
GFI 
0.979 RMR 
.001 
* 
p 
< .001

---

219
5.5.3 Preliminary Analysis – Product Quality Concern 
The product quality concern construct was measured using five items (PQC1 to PQC5). 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 5.4 for results). Evaluation of the 
correlation matrix through the KMO and Bartlett’s Test results in a high KMO statistic (.874) 
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 .86 to .93 and the variance 
explained was 81.07%. Cronbach’s alpha of .49 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 five items (retained from Step One) measuring product quality concern were subjected to 
CFA in Step Two (refer Figure 5.1) of the preliminary analysis. The results indicated a good 
fit for the measurement model. For example, 
χ
2
/df (1.451) was less than 5, GFI (.995), TLI 
(.996) and CFI (0.996) were above .90, RMSEA (.022) was less than .08 and RMR (.010) 
was less than .05. In addition, as displayed in Table 5.4, the standardized factor loadings of 
the four items ranged from .80 to .96, all exceeding the preferable criterion of .70 (Hair
 et al.
2006). The AVE was .75, exceeding the recommended criterion of .50 (Hair
 et al.
2006). The 
CR was .937, exceeding the threshold of .70 (Nunnally 1978). Taken together, items PQC1 to 
PQC5 were considered to constitute a reliable and valid measurement scale for this construct. 
As a result, all four items were retained for the overall measurement model. 

---

220