Electronic commerce research and applications: ECRA Co-Editors’ Introduction for Volume 9, Issue 6, November–December 2010
This issue offers a Research Directions article that explores issues in the application of structural equation modeling methods for IS and e-commerce research, and seven other Regular Research articles developed over the past two years on a variety of topics in e-commerce involving behavioral and design science research methods.
Our Research Directions article for this issue is ‘Lowers Bound on Sample Size in Structural Equation Modeling.’ The author is Chris Westland, one of the Co-Editors ofElectronic Commerce Research and Applications. It explores some issues with the application of structural equation modeling (SEM) for statistical analysis that comprises three classes of approaches. We often see it applied to the analysis of data in organizational and behavioral studies in IS and e-commerce research that involve surveys and multiple measures of related constructs. Westland argues that existing sample size heuristics – especially the general lack of good ones – have made it difficult for researchers to properly assess sample adequacy. This, in turn, has led to extensive under-sizing of sample sizes. His key observation is that it is inappropriate to work with SEM estimations based on an assumption that the requisite number of observations in the sample is a linear function of the number of parameters in the model. He contributes algorithms for computing minimum sample size that is needed to meet the stated SEM research objectives, showing that sample sizes ought to be a function of: (1) the ratio of indicator variable to latent variables in the model, and (2) power, significance and the minimum effect size. He provides additional insight by applying his algorithmic analysis to a large number of research articles in leading journals in the IS discipline. He reports that four-fifths of peer-reviewed published articles in the journals did not draw their statistical conclusions on the basis of sufficient sample size. He also points out that, among these published articles, the samples sizes tended to fall dramatically short of what he argues is necessary – average sample size was about one-half of that necessary to achieve statistical significance. This article has the potential to be an important contribution to our knowledge of SEM methods, and we expect this work to be relevant across a number of different academic disciplines other than IS, and to other research contexts beyond e-commerce.
The first of the Regular Research articles in this issue is by Bernd Skiera, Jochen Eckert and Oliver Heinz, who contributed ‘An Analysis of the Importance of the Long Tail in Search Engine Marketing.’ Search engine marketing relies upon the use of keywords to support consumer search for the products and services that they would like to buy. But how many are necessary to represent the ‘long tail’ of products that are offered on the Internet and demanded by consumers? The authors begin their research by challenging the conventional wisdom that there needs to be many keywords to support consumer search behavior. So then, perhaps an ‘80/20 rule’ is more appropriate? In fact, the authors find that the top 20% of all keywords are used in over 98% of all consumers’ searches. They also report that this same percentage of keywords is responsible for the generation of more than 97% of all subsequent click-throughs that consumers make, an indication of the efficacy of the top keywords. Thus, it is more like a ‘98/20 rule’ that seems to apply. Finally, they report that with no more than 100 keywords, it is possible to capture almost 89% of searches and 81% of click-throughs. These numbers are surprising, since they make search engine marketing seem less complex than one may expect. The authors established these results on the basis of an extraordinary data set of almost five thousand keywords and 10.1 million searches, and also report that the efficacy of the top 100 keywords tends to evolve over time. These empirical results are useful for e-commerce practitioners who are looking for guidance about where to focus their efforts to ensure their keywords are effective.
In spite of the capabilities that e-commerce firms have to scale their operations geographically via the Internet, and to eliminate distance as a major concern in their business model, consumers continue to express concerns about the geographical accessibility of the firms they do business with. The third article of the issue, by Sung-Eui Cho, addresses this issue and is entitled ‘Perceived Risks and Customer Needs of Geographical Accessibility in Electronic Commerce.’ The author suggests that the geographical accessibility of an e-commerce firm matters to a different extent based on the nature of products, services and market strategies of the firm. The theoretical model proposed involves the ease of delivery, the complexity of services, and the level of trust and reliability of the firm, which all affect the risk that a consumer perceives when ordering from a distant firm. Together these things determine the consumer’s needs for support to improve geographical accessibility. The author employed a survey of 129 South Korean respondents that was conducted in 2006. The empirical analysis offers evidence in support of the proposed theoretical framework.
The fourth article discusses the design of online auction mechanisms. ‘A Meta-Analysis on the Effects of Online Auction Options: The Moderating Effect of Value Uncertainty,’ by Yue-Wen Liu, Kwok-Kee Wei and Huaping Chen, explores the inconsistencies in what we know about how different design choices affect the outcomes of online auctions. They point to the importance of ‘value uncertainty’ as an explanatory moderating factor for the contradictions in the findings to date. They report several key results. First, a reserve price that is made public has a positive effect on the auction price, and is more influential when there is greater value uncertainty for an auction item. Second, when there is low uncertainty about an auction item, then having a secret reserve option will have a positive effect on the auction price of the item. Third, when the value of an auction item is of low uncertainty, then including a buy-it-now option will positively influence the likelihood of selling the auction item and support its price. The authors argue that the different design features – including the public reserve price, the secret reserve option and the buy-it-now option – all are important factors that lead to the outcomes of online auctions.
The fifth article ‘The Importance of Language Familiarity in Global Business e-Negotiation,’ by Hsiangchu Lai, Wanjung Lin and Gregory Kersten, makes a contribution in the area of electronic negotiation research. The language choice used in negotiation is known to be an important determinant of the outcomes that are observed. In this research, the authors developed an online experiment to evaluate the performance negotiators in text-based asynchronous systems, similar to those offered by conflict resolution web sites on the Internet. They evaluate the cognitive processes of the negotiation participants, by looking at objective measures of their behavior, including the number of offers and messages they send, and whether their negotiation results in transactional exchange. They varied the language of negotiation: either the negotiator’s native language or a foreign language is used. The authors find that language familiarity is a precursor of persuasion capability online, and that it supports their language and negotiation self-efficacy. They also report that the final outcome of the negotiation does not seem to be driven by language familiarity, although they note that negotiators who use a foreign language typically are less active negotiators. Also, buyers seem to be more affected by language choice than sellers.
The sixth article explores the issue of how online shopping recommender systems should be designed so they most effectively support e-retailing. In the ‘The Moderating Effects of Psychological Reactance and Product Involvement on Online Shopping Recommendation Mechanisms Based on a Causal Map,’ Namho Chung and Soon J. Kwon explore the interplay between quantitative and qualitative factors in establishing meaningful recommendations. They use a methodology called causal mapping, which emphasizes the set of constructs that are important to some kind of outcome, as well as the relationships among them. They also use an elaboration likelihood model (not to be confused with statistical likelihoods) to construct hypotheses about how consumers are affected by recommender systems. Their results indicate that the combined qualitative-quantitative approach offers additional power for helping consumers to achieve higher satisfaction and decision confidence, and increases their willingness to purchase.
Quality of service (QoS) continues to represent an important issue for operational performance in e-commerce. To be successful in selling online, Internet firms that offer their products and services need to be vigilant about the performance of their connections to the Web, their server operations, and their transaction completion capacity. When consumers experience frustrating response times, aborted transactions, and problems with data privacy, the firms they interact with are likely to lose sales and see diminished revenue flows. The seventh article, ‘RDRP: Reward-Driven Request Prioritization for E-Commerce Web Sites’ by Alexander Totok and Vijay Karamcheti, offers an economic interpretation that is evocative of Haim Mendelson’s 1985 treatment of queuing effects for the pricing of computer services. In that early work, Mendelson applied microeconomic analysis to gauge the marginal value of offering service computing cycles to the completion of queued jobs for which the willingness-to-pay of the service user could be discovered. In the present work, Totok and Karamcheti develop priority schedules for handling consumer transactions based on a value-focused approach – in this case, for Web service throughput. Their reward-driven request prioritization (RDRP) methodology has the goal of maximizing marginal profit and other beneficial outcomes from a provider’s service offerings. The authors write that their method predicts ‘future session structure by comparing requests seen so far with aggregate information about recent client behavior, and using these predictions to preferentially allocate Web server resources.’ To validate their RDRP method, the authors ran a set of experiments based on a benchmarking standard from the Transaction Processing Performance Council (www.tpc.org/tpcw) that is applicable for the Web server environment. The benchmark offers a transaction battery that is representative of a mix of transactions that might be what many companies would experience in a business environment that involves Internet-based selling of goods and services. This work is representative of some of the new and relevant interdisciplinary research directions that are associated with service science and the management of IT services.
The issue concludes with an article by Angel H. Crespo and Ignacio Rodríguez del Bosque. It explores ‘The Influence of the Commercial Features of the Internet on the Adoption of E-Commerce by Consumers.’ The authors identify a set of attributes that influence Internet shopping behavior and e-commerce, propose a model for how these features influences intention to conduct online purchases, and evaluate how consumer social demographics influence online purchasing behavior. They use Azjen’s theory of planned behavior, which emphasizes individual attitudes, normative beliefs, and perceived behavioral control, to explore product perceptions, perceived risks, and the nature of the shopping experience to evaluate adoption. The authors’ research was conducted in the context of a qualitative interview study of ten experts on the Internet in Spain, which later supported the development of a survey that yielded 998 valid responses. The authors find that attitudes towards e-commerce, subjective norms and perceptions of risk are the primary drivers of purchase decisions in online retailing.
We thank the authors and the reviewers for their service. This includes some new contributors from the statistics research community who offered pivotal assistance with the development of Chris Westland’s structural equation modeling article. We acknowledge Edward Bibo of Elsevier in Amsterdam, for his contributions over the past year. Edward recently stepped down from his role as the Journal Manager for ECRA. During his tenure, he made the operations of this journal run smoothly on the Elsevier side. We welcome Salai Birlasekar in Chennai, who joined our team as the new Journal Manager, and appreciate the continuing support of ECRA’s publisher, Rebecca Wilson, in the United Kingdom.
This has been a good year for ECRA. In 2010 we showcased special issues on ‘Social Networks and Web 2.0’ (January–February), ‘Theoretical and Empirical Advances in Electronic Auctions’ (March–April), ‘Nomadic Computing’ (May–June), and ‘Strategy, Economics and E-Commerce’ (September–October). We are grateful to this year’s Guest Editors: Julie Smith-David, Bin Wang and Chris Westland; Roumen Vragov, Chuck Wood and Claudia Loebbecke; Kemal Altinkemer; and Eric Clemons, Rob Kauffman and Thomas Weber. Their contributions of time and effort have resulted in high quality research articles for the journal.
The next issue in January–February 2011 will mark the beginning of ECRA’s 10th year of publication. We will be kicking off the year with a second guest-edited special issue on electronic auctions, and have some others in the works. The journal’s pipeline is solid as of this writing, with four issues of about eight articles apiece already slotted for publication. The journal is now receiving on the order of 260 submissions each year, of which about 17–18% are accepted. We think this is an appropriate and healthy acceptance rate for ECRA. We plan to maintain it as the number of submissions to the journal grows. We are seeing initial submissions of higher and higher quality, so acceptance for publication will necessarily become more competitive, even with a steady state acceptance rate. At any time, we have about 70–90 papers in the reviewing system in the process of review and development. The editorial team of ECRA continues to emphasize ‘developmental reviewing,’ to ensure that authors who submit articles with high potential to make innovative contributions and offer new knowledge obtain the support they need to be successful.
We are actively soliciting contributions from academic and industry colleagues who would like to submit Research Directions articles or develop special issues. So if you have a Research Directions idea, or would like to suggest a new special issue topic that the journal should explore, please contact one of the Co-Editors, so we can help you to shape it.
Information Systems and Management
Electronic Commerce Research and Applications
KAUFFMAN, Robert John; Chau, P.Y.K.; Payne, T.R.; and Westland, J.C..
Electronic commerce research and applications: ECRA Co-Editors’ Introduction for Volume 9, Issue 6, November–December 2010. (2010). Electronic Commerce Research and Applications. 9, (6), 473-475. Research Collection School Of Information Systems.
Available at: http://ink.library.smu.edu.sg/sis_research/2826
Copyright Owner and License