Understanding consumer online shopping behaviour from the perspective of transaction costs

2.2.4 Comparison between TAM and TPB 
A group of studies have compared TAM and TPB 
(Taylor and Todd 1995c, Chau and Hu 
2001a, Riemenschneider
 et al.
2003, Lin 2008, Huh
 et al.
2009)
. For example, Mathieson 
(1991) compared TAM and TPB in terms of how well they predict an individual’s intention 
to use an information system and stated the following: “
Both TAM and TPB predicted 
intention to use an IS quite well, with TAM having a slight empirical advantage; TAM is 
easier to apply, but only supplies very general information on users’ opinions about a system; 
TPB provides more specific information that can better guide development”
(Mathieson, 
1991, pp. 186). In line with Mathieson (1991), Hansen et al. (2004) tested both TRA and TPB 
in the context of online shopping, and found that TPB provided the better explanation to 
online consumer behaviour than TRA did. Similarly, Lin (2008) found that TAM explains 41 
per cent of the variance in behavioural intention, while TPB explains 46 per cent. 
The prior literature has revealed three main differences between TAM and TPB (Mathieson 
1991, Hansen
 et al.
2004). Firstly, TAM assumes that beliefs about PU and PEOU are always 

---

28
the primary determinants of use decisions, that is, these two beliefs are general regardless of 
the usage context (Lin 2008), while TPB asserts that beliefs are specific to each usage 
situation (Ajzen and Sheikh 2013). In some situations there may be other variables besides 
PU and PEOU that would well predict intention. In the e-commerce context, consumer 
behaviour is not only influenced by PU and PEOU, but also shaped by other important 
constructs such as trust, privacy concerns and hedonic values (Wu
 et al.
2012, Meskaran
 et al.
2013). On this point, TPB can provide a more accurate explanation to online consumer 
behaviour than TAM does.
The second difference is that TAM does not include social norm. In TPB, the social norm is 
an important construct to capture unique variance in intention (Ajzen 2011, Ajzen and Sheikh 
2013). It plays an inevitable role in determining online consumer behaviour since consumers 
usually follow the recommendations from their relatives, friends and people who share the 
same social value when they make purchase decision on certain brands (Manning 2009, 
Behjati and Pandya 2012).
Thirdly, TPB introduces PBC to take into account the effect of behavioural control. PBC is 
assumed to reflect past experience as well as anticipated impediments and obstacles (Ajzen 
2011). Pavlou and Fygenson (2006) pointed out that “most e-commerce studies follow TAM, 
implicitly assuming that behaviour is volitional. However, 
online shopping is indeed low in 
volitional control and
online consumers face several new constraints, such as the impersonal 
nature of the online environment, the extensive use of IT, and the uncertainty of the open 
Internet infrastructure (Al-Swidi
 et al.
2012). These issues call for the inclusion of PBC in e-
commerce adoption models, implying the superiority of TPB over the TAM.

---

29