Exploring the Drivers of E-Commerce through the Application of Structural Equation Modeling

Transcription

1 Exploring the Drivers of E-Commerce through the Application of Structural Equation Modeling Andre F.G. Castro, Raquel F.Ch. Meneses and Maria R.A. Moreira Faculty of Economics, Universidade do Porto R.Dr. Roberto Frias, s/n, Porto, Portugal {raquelm, Abstract. E-commerce is a form of trade that has gained increasing attention from consumers and sellers. However, despite high growth rates, e-commerce still has low levels of consumers. This study aims to determine the factors that influence the purchasing decision in e-commerce, in order to better understand acceptance or rejection of e-commerce among consumers. To this end, a first framework was constructed based on previous research on consumer adoption of e-commerce. Then, three categories (derived from interviews) were added. This more complete model was tested using the structural equation model based on partial least squares. The results obtained allow us to conclude that the perceived relative advantage, the ease of use, the drawbacks associated with a non-european Union country of sale and the perceived risk, directly influence the consumer s purchase intent in e-commerce. We also found a set of twelve variables that act as indirect influences. Keywords: e-commerce; influencing factors; drivers/determinants. 1 Introduction With the development and the widespread use of the Internet, the way people shop has changed, as has the way companies offer their products. Given the high growth rates of e-commerce consumerism, it is important to understand the dynamics of e- commerce and the factors that affect the choice of this purchase channel. The main aim of this study is to determine, through the development and application of an analytical model, which factors can influence (or influence the most) purchasing decisions in e-commerce. This research focuses on building a comprehensive model that portrays the consumer decision to purchase online. We built such a framework based on previous research on consumer adoption of e- commerce ([1], [2], [3], [4], [5], [6], [7]). A mixed methodology was employed. First, a qualitative study was conducted (adapted from the Grounded Theory) based on exploratory interviews, aimed at determining the specific factors that may influence the purchase decision in e- commerce. The results obtained with the exploratory interviews were integrated into the model resulting from the literature review and the subsequent quantitative test. The model estimation was performed in accordance with the Structural Equation Modeling of Partial Least Squares - PLS. Following the introduction, this paper is divided into four sections. The second presents a literature review. In the third section, the methodology are presented,

2 followed by the results. Finally, the last chapter presents the key findings and the study s limitations. 2 Literature Review This study will use the definition by [8], in that e-commerce corresponds to transactions that take place via the internet, involving the purchase and sale of goods and services that are delivered offline, as well as products that can exist and be delivered in digital form directly to the buyer online. Three approaches serve as the theoretical basis for the development of our research model - the consumer acceptance model of e-commerce, the understanding and mitigating uncertainty model, and the decision model for electronic commerce. Each of them provides some factors that may affect consumers in different contexts. We selected specific factors from each of the approaches as they fit within the context of e-commerce: buy, intention to buy, ease of use, perceived relative advantage, information security, privacy of information, information quality, seller reputation, social presence, pick up options, convenience, access to information, lower prices and customization. None of them would provide all the factors required to develop an in-depth model of the factors affecting e-commerce. However, those factors selected allowed us to formulate a number of hypotheses to test: H1: The intent of a consumer to purchase via e-commerce positively affects the buying decision through that channel. H2: The ease of use of the internet positively affects the consumer's purchase intention via e- commerce. H3: Perceived Risk by consumers negatively affects Intent to Purchase via e-commerce. H4a: Consumer confidence positively affects the consumer's intention to buy via e-commerce. H4b: Consumer confidence negatively affects the consumer s perceived risk of a transaction via e-commerce. H5: Perceived relative advantage positively affects the intention to buy via e-commerce. H6: Concerns with information security increase the perceived risk in e-commerce. H7: Concerns about the privacy of information increase perceived risk in e-commerce. H8a: The information quality decreases perceived risk in e-commerce. H8b: The information quality increases consumer confidence in e-commerce. H9: A seller s (positive) reputation positively affects consumer confidence in e-commerce. H10: The social presence of a seller positively affects the consumer s confidence in e- commerce. H11: The perception of increased options in e-commerce positively affects the relative advantage perceived in this channel. H12: The perception of convenience in e-commerce positively affects the perceived relative advantage of online shopping. H13: The perception of getting more and better access to information in e-commerce positively affects the perceived relative advantage of using this channel. H14: The perception of getting a low price in e-commerce positively affects the perceived relative advantage of purchasing decisions in this channel. H15: The perception of possible product customization in e-commerce positively affects the perceived relative advantage of using this channel.

3 3 Methodology To do this research, a mixed approach was applied. Thus, in a first phase, an exploratory methodology was used, based on an in-depth analysis of interviews, to identify possible new variables that influence consumption in e-commerce, which could be included in the model, following accordingly, mutatis mutandis, the methodology of the Grounded Theory [11]. To this end, semi-structured interviews were conducted, between 1 st and 11 th April at the University of Porto. The data was collected, analyzed, sorted and categorized into concepts that emerged from the data. Later, these concepts gave rise to the properties or subcategories, establishing relationships among them. In a second phase, a quantitative analysis was performed in order to test the model resulting from the theoretical analysis and the exploratory survey. In order to empirically test the final model resulting from the literature review and the exploratory interviews, a research questionnaire was prepared, involving the collection of a large amount of information related to the target population. The questionnaire was made up of 35 questions, in both multiple choice and scale evaluation. The application of the questionnaire to potential respondents was conducted via , where the cooperation of all students of the University of Porto was requested and the purpose of the questionnaire explained. The questionnaire was sent on 20 th June 2011, to the entire student population of the University of Porto, and it was available until 31 st July. Given the impossibility of studying the entire population of e-consumers, we decided to focus this study on higher education students. As described by [10], several studies have used them as an object of study, since it is recognized that students are a useful proxy for characterizing online consumers. Thus, the survey of the available research was performed, and the PLS was set up, as an estimation method for the model, in order to ascertain the results of the research and to draw conclusions. 4 Results 4.1 Results from the Qualitative Analysis The interviews identified a set of enhancing variables that influence consumer intention to buy online, which were not part of the theoretical model. The list of all variables identified is as follows: shipping costs, urgency of purchase, recommendation from friends, product experimentation requirements, disadvantages associated with the extra-eu countries, widespread use of e-commerce as a way to buy and success of previous experiences. Only the variables which had been referenced more than three times (more than 50% than the maximum value) since the others could constitute circumstantial and non-significant phenomenon, were included in the model. Thus, the data from the interviews yielded the following variables to be added to the proposed model: Product experimentation requirements - within the variable perceived risk, it aims to demonstrate that, in products where there is a high need for experimentation thereof,

4 consumers feel they are faced with a transaction of increased risk, if they choose to buy them via e-commerce; Disadvantages associated with the extra-eu countries this variable tends to directly influence consumer intention to buy from non-eu countries, presenting some drawbacks to online shopping, such as the delay in product delivery, possibility that products be retained at customs and thus require the payment of customs fees, the possible need for payment in currencies other than the consumer s, etc.. Urgency of purchase - within the variable perceived relative advantage, given the time difference between the act of purchase and the act of delivery, consumers may not find it advantageous to purchase via this channel. When they have urgent needs, they resort to traditional retail. Accordingly, the following hypothesis can be presented: H16: The product experimentation requirements increase the perceived risk in e-commerce. H17: The disadvantages associated with extra-eu countries negatively affect consumer intention to buy via e-commerce. H18: The urgency of purchase negatively affects consumer buying intention via e-commerce. Schematically the new proposed model, taking into consideration the above factors (light), and the factors from literature review (dark) is presented in Figure 1. Experiment Requirement Security H16 H6 Disadvantage Extra-EU Ease of Use H17 H2 Privacy of Quality Seller Reputation Social Presence Pick Up Options H7 H8a H9 H10 H8b H11 H12 Convenience Perceived Risk Trust Perceived Relative Advantage H13 H4b Access to H3 H4a H5 H14 H15 H18 Lower Prices Figure 1. Analysis model Intention to Buy H1 Urgency of Purchase Customization Buy

5 4.2 Results from the Quantitative Analysis The number of valid responses to be statistically analyzed was questionnaires with which constituted our database. We checked the reliability and convergent and discriminant validity of the constructs. To do so, we used some indicators: Cronbach's alpha: all variables in our model, the Cronbach alpha have a value greater than 0.6, the minimum acceptable, according to [13] and [14];have urgent needs, they resort to traditional retail. Fornell's composite reliability: all model variables have a Fornell confidence value higher than the acceptable one, 0.7 ([10]); AVE (average variance extracted): in the model all variables, have a value greater than the minimum value of 0.50 [14]; The square root of the AVE of each variable must not be less than the correlations of this variable with the other ([15]), which also occurs in the model in question; Correlations between all the variables should be less than 0.90, as indicated by [9], which is also confirmed in this model; Having secured the validity and reliability of measurement models, the structural model estimation was performed via PLS using the SmartPLS software. In order to decide whether to accept or reject a hypothesis, for a significance level of 5%, the value of 1.96 for the T-Statistic should be obtained as the reference. At a significance level of 5%, all the assumptions set out in the model, with the exception of hypotheses H4a (T-Statistic=1.7231<1.96) and H10 (T- Statistic=1.1934<1.96), can be accepted. Regarding hypothesis H4a (consumer confidence positively affects the purchase decision via e-commerce), in the study population, it was not possible to prove the statistical significance of the variable confidence in consumer purchase intent. Although this relationship has been validated in previous studies, here, with the data collected, we cannot validate that trust has a direct influence on the consumer's intention to buy via e-commerce. It was also not possible to confirm H10 (the social presence of a seller positively affects consumer confidence in e-commerce) in the population under study; we cannot say that the effort of e-commerce business sellers to equip their sites with systems enabling greater social contact between buyers and sellers, via chat, online customer support, etc.., has a positive impact on consumer trust in the e-commerce universe. Therefore, taking these two assumptions of the model in question, since they could not be accepted, the final results of estimation model are presented in Figure 2, with the coefficients associated with each hypothesis under consideration, and the R 2.

6 Experiment Requirement Security Disadvantage Extra-EU Ease of Use Privacy of Quality Seller Reputation Perceived Risk R 2 =0.47-0,26 Trust R 2 = Intention to Buy R 2 =0.41 0,49 Buy R 2 =0.24 Pick Up Options Convenience Perceived Relative Advantage R 2 = Access to Lower Prices Urgency of Purchase Customization Figure 2. Final results of estimation model 5 Conclusion In this study, we analyzed the factors that influence the decision to purchase via e- commerce, thus intending to contribute to a broader understanding of the online shopping phenomenon. We found that there are a set of four variables that influence directly the consumer s purchase intention through e-commerce: perceived relative advantage, ease of use, disadvantages associated with the extra-eu countries and perceived risk. The first two factors act as incentives to purchase in e-commerce, where perceived relative advantage exerts greater influence. Conversily, the variables disadvantages associated with the extra-eu countries and perceived risk tend to decrease the intention to use the internet to shop, working as obstacles to online consumption. This study also found some limitations that should be mentioned. The sample used in the study may not be representative of all consumers. Although students are the group that most often purchases online and are also the target of many research studies on this matter, the conclusions must be relativized to the sphere of study. Hence, it may not be possible to generalize the findings to other groups of consumers. As avenues for future research, a study could be conducted on different product categories in order to assess if the influence of the factors identified remained the same or vary according to the specifics of the products concerned.

7 Acknowledgement This work is funded by the ERDF through the Programme COMPETE and by the Portuguese Government through FCT - Foundation for Science and Technology, project PTDC/EGE-GES/099741/2008. References 1. Chen, Y.-Y. (2012) Why Do Consumers Go Internet Shopping Again? Understanding the Antecedents of Repurchase Intention Journal of Organizational Computing and Electronic Commerce, Vol. 22, Nº 1, pp Guo, Li (2011) A Research on Influencing Factors of Consumer Purchasing Behaviors in Cyberspace, International Journal of Marketing Studies, Vol. 3, No. 3, pp Kamarulzaman, Y. (2011) E-consumer behaviour: Exploring the drivers of e-shopping, European Journal of Social Sciences, Vol. 23, Nº 4, pp Zhou, L., Dai, L., Zhang, D. (2007) Online shopping acceptance model-a critical survey of consumer factors in online shopping, Journal of Electronic Commerce Research, Vol.8, Nº, pp Monsuwé, T., Dellaert, B. and Ruyter, K. (2004) What drives consumers to shop online? A literature review, International Journal of Service Industry Management, Vol. 15 No. 1, pp Park, C., Jun, J.-K.(2003) A cross-cultural comparison of Internet buying behavior: Effects of Internet usage, perceived risks, and innovativeness, International Marketing Review, Vol. 20, No. 5, pp Moore, R. and Breazeale, M. (2010). Electronic Commerce Research: The first 15 years in the fields of Marketing, Management, and Systems, Marketing Management Journal, Vol. 20, pp Coppel, J. (2000), E-commerce: Impacts and Policy Challenge, Economics Department Working Papers, Nº 252, pp Pavlou, P., Liang H. and Xue, Y. (2007), Understanding and Mitigating Uncertainty in Online Exchange Relationships: A Principal-Agent Perspective, MIS Quarterly Vol. 31, Nº. 1, pp Kim, D., Ferrin, D. and Rao, H. (2008), A trust-based consumer decision-making model in electronic commerce: The role of trust, perceived risk, and their antecedents, Decision Support Systems, 44, pp Glaser, B. and Strauss A. (1967), The discovery of grounded theory. Chicago: Aldine. 12. Lu, C.-S., Lai, K.-h. and Cheng, T.C.E. (2007) Application of structural equation modeling to evaluate the intention of shippers to use Internet services in liner shipping, European Journal of Operational Research, Vol. 180, Nº 2, pp Sekaran, U. (2005), Research methods for business a skill building approach (4th edition), New York: John Wiley and Sons, 4ª Ed. 14. Hair, J., Anderson, R.. Tatham, R. and Black, W. (2006), Multivariate Data Analysis, Prentice Hall. Upper Saddle River: Ed. Prentice Hall, 5ª Ed. 15. Fornell, C. and Larcker, D. (1981), Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, Vol. 18, pp: