Understanding consumer online shopping behaviour from the perspective of transaction costs

4.2 RESEARCH METHODS 
 
Researchers, as they prepare their studies, can choose quantitative, qualitative, or mixed 
methods. Qualitative method is primarily concerned with an in-depth exploration of the 
subject under study and focuses on finding and examining as many details as possible. 
Quantitative method is chosen when the purpose of the research is to collect primary data, 
that is, data gathered and assembled specifically for the study, as opposed to secondary data 
(Babbie 2007). The value of the quantitative method is its ability to test hypotheses, compare 
responses, and produce generalizations. The ability to test hypotheses also allows the 
relationships in theories to be tested and validated (Babbie 2007).

---

165
Previous TC research has used quantitative methods through survey investigation to analyse 
the relationships between constructs of interest (Liang and Huang 1998, Teo
 et al.
2004, Kim 
and Li 2009b, Yen
 et al.
2013, Wu
 et al.
2014). In particular, Teo et al. (2004) tested a 
consumer TC model of online shopping in U.S.A and China using structural equation 
modelling of survey data. In line with that and other studies, the present study focuses on 
theory testing at the consumer level of analysis and adopts a primarily quantitative and 
survey-based method. It is argued that the quantitative method is appropriate because it 
allows for the collection of answers to a number of questions from a large sample of the 
population of online shoppers, which is required to achieve valid results (Wu
 et al.
2014). 
The survey questionnaire approach is the most common method of generating and collecting 
primary data and is considered to be the best method available to social scientists interested 
in collecting data for describing a population too large or too dispersed to observe directly 
(Babbie 2007). This type of research method asks respondents to answer on a Likert-type 
scale along a continuum (Saris and Gallhofer 2007). Items on the survey are assigned a value, 
for example, “strongly disagree” (value of 1) to “strongly agree” (value of 7). The advantages 
of a questionnaire are economy, speed, lack of interviewer bias, and the possibility of 
anonymity and privacy to encourage responses that are more candid. Survey results can be 
quantified to provide additional empirical support (Babbie 2007). 
In the present study, closed questions in a survey-based format were used to address research 
questions and hypotheses. This method was selected for the following reasons. Firstly, the 
study intends to test the TCT in the context of online shopping by gathering a wide range of 
information from a large population, specifically it aims to explicate the relationships 
between consumer TCs and their online behavioural consequences, and to identify the 

---

166
antecedents of consumer TCs associated with online purchasing from a large sample of the 
population of online shoppers. Accordingly, a survey is appropriate to accomplish this goal of 
the study.
Secondly, given that limited amount of data available from published sources, the use of a 
structured, self-reported survey for the purposes of data collection is both legitimate and 
dominant in consumer behaviour research (Ganesh
 et al.
2010, Hasan 2010). Primary data 
was collected by the researcher because the type of data required for this research is 
customers’ perceptual data in the context of online shopping in China which were not 
available in any public sources. A survey method also enables data to be confidentially and 
inexpensively collected from a targeted but geographically diverse population, thereby 
making it suitable for collecting the data from the study sample that is used to test the 
proposed model in this study.
Thirdly, a survey with closed questions allows for standardization of the data collected across 
individuals and provides a rapid turnaround time for data collection. Responses can be easily 
classified, thus making analysis very straightforward (Davis and Cosenza 1895). Fourthly, the 
data required for this study are perceptual, and consumers’ perception is the reflection of 
people’s belief and attitudes towards online consumption experience, while the survey 
method is the best to serve such purposes (Kerlinger 1973). Previous studies investigating 
TCs have shown that a survey approach is an efficient and acceptable data collection method 
to adopt (Yeh
 et al.
2012b, Yen
 et al.
2013, Wu
 et al.
2014).
Following the previous research, 
the study used survey data to test the hypotheses. 

---

167