Revise your research topic based on the feedback received from the last assignment on the Research Topic. Now you are to prepare a literature review using only research articles from MIS Quarterly to justify the important of this research topic.
This 30 page APA paper should end with a diagram of the research model this is supported by at least 40 references in your reference list that you cite in this paper ( 20 references added below)
22) Grover, Varun; Ramanlal, Pradipkumar.(Dec1999). Six myths of information and markets: Information technology Networks, Electronic Commerce and the Battle for Consumer Surplus. Vol. 23 Issue 4, p465-495. 31p. 8 Diagrams. DOI: 10.2307/249486
23) El Sawy, Omar A.; Malhotra, Arvind; Gosain, San Jay; Young, Kerry M. (Sep 1999), IT-Intensive Value innovation in the electronic economy: Insights from Marshall Industries Vol. 23 Issue 3, p305-335. 31p. 2 Black and White Photographs, 7 Diagrams, 3 Charts, 2 Graphs. DOI: 10.2307/249466.
24) Choudhury, Vivek; Hartzel, Kathleen S.; Konsynski, Benn R. (Dec1998). Uses and Consequences of Electronic Markeys : An Empirical Investigation in Aircraft Parts Industry. Vol. 22 Issue 4, p471-507. 37p. 3 Diagrams. DOI: 10.2307/249552.
25) Leong, Carmen., Pan, Shan L., Newell, Sue., Cui, Lili. (2016). The emergence of self-organizing e-commerce ecosystems in remote villages of china: A tale of digital empowerment for rural development. MIS Quarterly, 40(2), 475-483.
26) Duan, Wenjing., Gu, Bin., Whinston, Andrew B. (2009). Informational cascades and software adoption on the internet: An empirical investigation. MIS Quarterly, 33(1), 23-48.
27) Yulin Fang., Israr Qureshi., Heshan Sun., McCole, Patrick., Ramsey, Elaine., Kai H. Lim. (2014). Trust, Satisfaction, and Online repurchase intention: The moderating role of perceived effectiveness of e-commerce institutional mechanisms. MIS Quarterly, 38(2), 407-424.
28) Kil-Soo Suh., Young Eun Lee. (Dec 2005). The effects of virtual reality on consumer learning: an empirical investigation. MIS Quarterly,29(4),673-697. DOI: 10.2307/25148705
29) Hong, Weiyin; Hess, Traci J.; Hardin, Andrew. (Jun2013),When Filling the wait makes it feel longer : A Paradigm Shift Perspective for Managing Online Delay. Vol. 37 Issue 2, p383-A9. 33p. . DOI: 10.25300/MISQ/2013/37.2.04.
30) VanderMeer, Debra., Dutta, Kaushik., Datta, Anindya. (June 2012). A cost-based database request distribution technique for online e-commerce applications. MIS Quarterly,36(2),479-507. DOI: 10.2307/41703464.
31) Lee, Dong-Joo., Ahn, Jae-Hyeon., Bang, Youngsok. (June 2011). Managing consumer privacy concerns in personalization: a strategic analysis of privacy protection. MIS Quarterly,35(2),423-A8 .DOI: 10.2307/23044050.
32) Bose, I., & Leung, A. C. M. (2019). Adoption of identity theft countermeasures and short- and long-term impact on firm value. MIS Quarterly, 43(1), 313–327. DOI: 10.25300/misq/2019/14192
33. Li, H., Fang, Y., Lim, K. H., & Wang, Y. (2019). Platform-based function repertoire, reputation, and sales performance of e-marketplace sellers. MIS Quarterly, 43(1), 207–236. DOI: 10.25300/misq/2019/14201
34) Lee, A. (1999). Six myths of information and markets technology: Information technology networks, electronic commerce, and the battle for consumer surplus. MIS Quarterly, 23(4).
35) Vivek, C., Kathleen, S. H., and Benn R. K. (1998). Uses and Consequences of Electronic Markets: An Empirical Investigation in the Aircraft Parts Industry. MIS Quarterly, 22(4), 417-417.
36) Wang, W., Benbasat, I., (2009). Interactive decision aids for consumer decision making in e-commerce: The influence of perceived strategy restrictiveness. MIS Quarterly, 33(2), 293-320
37) Komiak, Sherrie Y. X.; Benbasat, Izak. (2006). The Effects of Personalization and Familiarity on Trust and Adoption of Recommendation Agents. MIS Quarterly, 30(4), 941-960.
38) Chatterjee, Debabroto; Grewal, Rajdeep; Sambamurthy, V. (2002). Shaping Up for E-Commerce: Institutional Enablers of The Organizational Assimilation of Web Technologies. MIS Quarterly, 26(2), 65-89.
39) Wetzels, Martin, Odekerken-Schröder, Gaby, van Oppen, Claudia. (2009). Using PLS Path Modeling for Assessing Hierarchical Construct Models: Guidelines and Empirical Illustration. MIS Quarterly, 33(1), 177-195.
40) Shun Ye; Guodong (Gordon) Gao; Viswanathan, Siva. (Dec2014) STRATEGIC BEHAVIOR IN ONLINE REPUTATION SYSTEMS: EVIDENCE FROM REVOKING ON EBAY1 MIS Quarterly. Dec2014, Vol. 38 Issue 4, p1033-1056. 24p. 1
41) Etzion, H., and MinSeok, P. (2014). Complementary online services in competitive markets: Maintaining profitability in the presence of network effects. MIS Quarterly, 38(1), 231-247.
42) Hibbeln, M., Jenkins, J., L., Schneider, C., and Valacich, J., S. (2017). How is your user feeling: Inferring emotion through human computer interaction devices. MIS Quarterly, 41(1). 1-21.
E-Commerce: Downfall of Brick and Mortar Stores 1
E-Commerce: Downfall of Brick and Mortar Stores 7
Down Fall of Brick and Mortar Stores
Downfall of Brick and Mortar Stores
E-commerce is growing at a higher rate across the world as people begin to adopt online shopping as opposed physical shopping. In 2018 for instance, consumers in United States spent $517.36 billion on online shopping –accounting for 14.3 percent of the total retail sales. This presents a 15 percent growth from the total e-commerce sales recorded in 2017 amounting to $449.88 billion. E-commerce giants like Amazon continue to register a tremendous growth –commanding a 40 percent of the total online retail in U.S. Due to this unprecedented growth, e-commerce is expected to dominate the retail sector to become the largest retail channel globally by the year 2020. This new wave has begun to sweep some major brick-and-mortar retailers that rely on physical stores to sell their products and services. Failure by management to utilize technology in shifting their operations to online e-commerce has led to their agonizing downfall. This paper seeks to conduct a systematic analysis regarding the rise of e-commerce and its impact to the downfall of major brick-and-mortar stores witnessed over the last few years. Additionally, the paper will extend focus on other factors that may have led to the downfall of renowned physical retail stores across the United States.
Part I: Introduction
Remarkably, it is beyond no doubt that brick-and-mortar stores are facing a lot of pressure resulting from growth in e-commerce. Failure to embrace technological shift to e-commerce is forcing major brick-and-mortar retailers out of the market. Physical stores are “being pushed towards the path of extinction” for failing to adapt to new changes in the market. The year 2017 and 2018 marked the biggest trends in the closure of key reputable physical retail stores due to the “apparent disaster of decline in sales.” Macy’s, J.C. Penney, Sears, Bon-Ton, Fitch, Toys “R” Us, and Abercrombie, just to mention a few, are some of giant physical retailers which have been forced to close down their stores.
Truly, we are experiencing heightened level of big retail closures that has hit the U.S. in the last 10 years. On the other hand, companies like Amazon that have embraced e-commerce continue to register massive growth. In 2018, Amazon hit $206.82 billion in sales from their e-commerce site Amazon.com. Customers have developed confidence with companies which have e-commerce apps and websites since they offer them an opportunity to shop at their own convenience. By 2020, the growth in e-commerce is expected to hit a historical 200 percent –meaning that customers will be continue to find it convenient to shop online since it saves them time required to visit physical stores and make manual selection of items.
Part II: Literature Review
Many studies have been conducted across the world as researchers seek to understand the impact of growth in e-commerce and its contribution to the downfall of brick-and-mortar stores. The studies covered in this section are aimed at establishing the role of rising tide in e-commerce as well as any other factor that may have resulted to the widespread downfall of giant physical retail stores across the U.S.
Growth in technology has offered a platform for businesses to expand their operations. Web technological platforms offer a powerful business resource –mainly the e-commerce. Thus, technological advancements have offered organizations an opportunity to make strategic investment by acquiring websites and apps that can enable customers to shop for products and services online (Chatterjee, Grewal & Sambamurthy, 2002). Effective web assimilation offers businesses an opportunity to succeed since the maintenance costs are less compared to those of physical stores. When developing e-commerce systems, developers need to follow a logical structure that unifies all the elements of the system since this determines their success. According to Fang, Lim, Qian & Feng (2018), “e-commerce has emerged as an important strategic resource for firms facing fierce competition in traditional markets.” Thus, brick-and-mortar stored are being forced to make strategic decision on the need to allocate resources required to expand their online presence. Notably, e-commerce companies provide consumers with “multiple points of contact, personalized information, and convenient transactions.”
Unlike brick-and-mortar stores that require a lot of resources to coordinate their operations including supply chain management and communication, e-commerce retailers are able to minimize these costs hence increasing their profit margin (Xiao & Benbasat, 2011). Additionally, e-commerce companies have a huge advantage over physical stores in many aspects. Ghose (2009) found out that sellers’ reputation is high in online markets compared to physical brick-and-mortar environment. Thus, is because consumers have the ability to review products and recommend them to others. Also, e-commerce helps to minimize hierarchical structures hence making management easy and effective. Moreover, it is easy to search for a product and services while their prices are affordable due to reduced logistics.
E-commerce business model creates a more flexible, dynamic, successful, and long-lasting relationship between buyers and sellers. This relationship is created through easy communication between online retailers and consumers. Unlike in brick-and-mortar retailers whereby a customer must visit the stores to interact with the seller, e-commerce offers a feedback portal that allows consumers to contact them anytime and receive real-time responses (Mithas, Jones & Mitchell, 2008). This makes e-commerce companies to build strong customer trust and loyalty. E-commerce stores are definitely more user-friendly compared to brick-and-mortar retail outlets. According to Webster & Ahuja (2006), e-commerce sites are more engaging allowing buyers to easily navigate through them and access all the necessary information. The “help” buttons on e-commerce websites and apps helps consumers link up with support team and access quick help regarding the company’s products and services.
Notably, online e-commerce sites generate income by applying different revenue models. The two popular models used by e-commerce companies is brokerage model and advertising model. eBay.com for instance uses brokerage model whereby sellers pay the company for every transaction they make through their website. On the other hand, Taobao.com offers sellers a platform to advertise their products and services and thus pay an advertising fee (Chen, Fan & Li, 2012). However, sellers prefer advertising model due to its exposure and the ability to generate more social welfare. The adoption of e-commerce model by consumers in depends entirely on a number of factors. Notably, consumers getting information regarding a product and use it to make decisions on whether to purchase product. It was found out that “technological characteristics (download delay, Website navigability, and information protection), consumer skills, time and monetary resources, and product characteristics (product diagnosticity and product value) influence the customer purchasing behavior” (Pavlou & Fygenson (2006). E-commerce companies strive to improve operations based on the current business frameworks being applied in the world of online transactions. Continuous improvements ensure that e-commerce companies enhance their effectiveness hence attracting unexpected segments of new customers (Albert et al., 2004). These endless improvements make e-commerce companies stand out ahead of brick-and-mortar retailers. Notably, e-commerce retailers continue to experience new and promising opportunities. The growth of social media platforms has been a big boost to e-commerce companies. Integrating social media to a company’s online marketing model can help to increase product awareness hence driving more sales. Based on a study involving Groupon.com, Li & Wu (2018) found out that “social media plays a key role in spreading the information regarding a product hence increasing awareness and demand.”
E-commerce companies engage in market research aimed at collecting lot of data regarding their potential customers. This data is subjected to data-based analytical methods in order to establish the purchasing behaviors of customers as well as their needs and wants (Padmanabhan, Zheng & Kimbrough, 2006). The availability of online data collection and analytical techniques makes it easy for e-commerce retailers to collect and analyze consumer data. E-commerce retailers spend less time and resources in collecting and analyzing huge amounts consumer data compared to brick-and-mortar stores that spend a lot of time and resources to conduct market research. Xiao & Benbasat (2007) also strives to the techniques used by online retailers to study consumer purchasing behaviors and interests. Online retailers that rely on RA software to gauge the decisions that customers make before purchasing products online hence helping them to develop products which satisfy consumer needs and wants.
According to Lin, Goh & Cheng (2017), e-commerce sites also depend on product recommendation networks to drive demand for their products. This strategy has gained a lot of popularity among e-commerce retailers in the contemporary world of business. Based on a study involving Tmall.com, it was found that both co-purchase and co-view networks play a key role in increasing the product’s demand and awareness. According to Wu, Huang & Zhao (2019), products and services review is a very crucial element for e-commerce retailers. Notably, customers express their satisfaction or dissatisfaction through reviews whereby they give their honest opinions regarding products or services. Online retailers strive to engage customers carefully in order to satisfy their needs –leading to positive reviews. Negative reviews might discourage potential buyers from buying products or services from an online retailer. According to Mudambi & Schuff (2010), product reviews help to supplement the information provided by e-commerce retailers regarding their products and services. Notably, customers in brick and mortar stores have no access to products reviews unlike e-commerce sites whereby consumers find reviews which are helpful in guiding their purchasing decisions.
In modern days, e-commerce companies are facing a myriad of challenges which derail their operation and ultimate growth. E-commerce sites are unable to create online brand positioning due to low skills regarding search engine optimization (Dou, Lim, Su, Zhou & Cui, 2010). Thus, most of them lack the capacity to maintain competition and maximize sales online. Despite the fact e-commerce B2C business model has been operational for quite some time, the online environment uncertainties remain a major impediment. According to, Pavlou et al. (2007), “consumers are reluctant to engage in online exchange relationships hence presenting a primary barrier to online transactions.” Fear of seller opportunism, privacy issues, and online security concerns are some of the major problems that online retailers must address in order to win the trust of consumers. According to Tan & Canfetelli (2016), e-commerce systems experience technical failure that affect consumer shopping experience negatively. This may lead to customer dissatisfaction since their needs and requests might not be met. The study sought to explore the challenges facing e-commerce systems making them undesirable for some customers.
Tremendous growth in e-commerce has created a prime ground for deception and fraud that take place online. This trend calls for the attention of consumer organizations and government agencies to detect online merchants who are exploiting customers through e-commerce websites. Consumer education is also necessary to create awareness among consumers on how to avoid online deception (Koch & Schultze, 2011). According to Hinz et al., (2011), maintains that online retailers have been exploiting customers through discriminative prices which are exorbitant in nature. Prices set by online retailers are dynamic and in most cases, they are determined by “buyer’s willingness to buy.” For instance, Amazon was once found to have sold a DVD to one customer at $22.24 and another one at $22.74. Hence, despite their convenience, there prices still remain high compared to physical stores. In some cases, electronic systems are bound to fail hence affecting the organizational operations, and so is e-commerce systems.
Notably, technological shift to e-commerce has introduced a new B2C business model that has pushed traditional brick-and-mortar retailers to the wall. This wave is forcing physical stores to adopt an online business model in order to counter competition and avoid the path of extinction. However, there is need to understand that e-commerce businesses face a lot of challenges which make them undesirable to some consumers. Lack of privacy and security, data theft, fraud, and price discrimination are some of the issues that make consumers shy away from online shopping.
Notably, brick-and-mortar stores are still popular since consumers can interact with sellers and inquire about the products and services. Physical stores also provide consumers with “instant gratification when a purchase is made.” Brick-and-Mortar stores also offer consumers with firsthand shopping experience allowing customers to test products before purchasing them. Moreover, there are some products like cars which require offline selling. All these factors make brick-and-mortar retailers somehow popular.
Thus, the literature in the articles has not substantially linked the rise of e-commerce to the downfall of brick-mortar-stores in the American market. Thus, there is need to carry out further studies to answer this question: “Is it purely the technological shift to e-commerce that’s squeezing brick and mortar retailers, pushing them out of the market?”.
More is needed to understand the psychological and economic factors that influence the decision to buy online versus face to face.
For each article you need to provide a detailed discussion of the major constructs (factors) in that paper. Then build a table that show the constructs and description from each article.
Then you need to build a model to test your research question which will continue to change as you dig deeper into these research articles.
Grade B – this team is off to a good start – make the above changes in Red.
Albert, Terri C., Goes, Paulo B., Gupta, Alok. (Jun 2004). A model for design and management of content and interactivity of customer-centric web sites. MIS Quarterly, 28(2), 161-182. DOI: 10.2307/25148632
Chatterjee, D., Grewal, R., & Sambamurthy, V. (2002). Shaping up for E-Commerce: Institutional enablers of the organizational assimilation of web technologies. MIS Quarterly, 26(2), 65. DOI: 10.2307/4132321.
Chen. J, Fan. M, Li.M. (Sep 2016). Advertising Versus Brokerage Model for Online trading. 40(3). 575-601.
Dou, Lim, Su, Zhou, & Cui. (2010). Brand positioning strategy using search engine marketing. MIS Quarterly, 34(2), 261. DOI: 10.2307/20721427.
Ghose, A. (Jun 2009). Internet exchanges for used goods: an empirical analysis of trade patterns and adverse selection. MIS Quarterly, 33(2), 263-291. DOI: 10.2307/20650292.
Hinz, Oliver., Hann, II-Horn., and Spann, Martin. (March 2011). Price discrimination in E-Commerce? An examination of dynamic pricing in name-your-own price markets. MIS Quarterly, 35(1), 81-98. DOI: 10.2307/23043490.
Koch, Hope., Schultze, Ulrike. (2011). Stuck in the conflicted middle: A role-theoretic perspective on b2b e-marketplaces, MIS Quarterly, 35(1), 123-146. DOI: 10.2307/23043492.
Lin. Z, Goh. K, Cheng. S, (Jun 2017). The Demand Effects Of Product Recommendation Networks: An Empirical Analysis of Network Diversity and Stability. 41(2), 397-427.
Mithas, S., Joni L., Mitchell W. (Dec 2008). Buyer intention to use internet-enabled reverse auctions: the role of asset specificity, product specialization, and non-contractibility. MIS Quarterly, 30(4), 705-724. DOI: 10.2307/25148868.
Mudambi. S, Schuff. D, (Mar 2010). What makes a helpful online review? a study of customer reviews on amazon.com. MIS Quarterly, 34(1), 185-200. DOI: 10.2307/20721420.
Padmanabhan, B., Zhiqiang, Z., & Kimborough, S. (June 2006). An Empirical Analysis Of The Value Of Complete Information For ECRM MODELS. MIS Quarterly, 30(2), 247-267. https://doi.org/10.2307/25148730/MISQ. DOI: 10.2307/25148730.
Pavlou, Paul A., Fygenson, Mendel. (Mar 2006). Understanding and predicting electroni commerce adoption: an extension of the theory of planned behavior. MIS Quarterly, 30(1), 115-143. DOI: 10.2307/25148720.
Pavlou, Paul A., Huigang Liang., Yajiong Xue. (Mar 2007). Understanding and mitigating uncertainty inonline exchange relationships: a principal– agent perspective. MIS Quarterly, 31(1), 105-136. DOI: 10.2307/25148783.
Tan, C., and Cenfetelli, I., (2016). An exploratory study of the formation and impact of electronic service failures. MIS Quarterly, 40(1), 1-29.
Webster, J., & Ahuja, J.S. (Sept 2006). Enhancing the design of web navigation systems: the influence of user disorientation on engagement and performance. MIS Quarterly, 30(3),661-678. https://doi.org/10.2307/25148744/MISQ. DOI: 10.2307/25148744.
Wu, J., Huang, L., & Zhao, J. L. (2019). Operationalizing regulatory focus in the digital age: Evidence from an E-Commerce context. MIS Quarterly, 43(3), 745–764. DOI: 10.25300/misq/2019/14420.
Xiao, B., Benbasat, I. (2007). E-Commerce products recommendation agents: Use, characteristics, and impact. MIS Quarterly, 31(1), 137-209. DOI: 10.2307/25148784.
Xiao, Bo., Benbasat, Izak. (2011). Product-related deception in e-commerce: A theoretical perspective, MIS Quarterly, 35(1), 169-196. DOI: 10.2307/23043494.
Xitong L, Lynn W. (Dec 2018). Herding and social media word-of-mouth: evidence from groupon. 42(4), 1331-1351. DOI: 10.25300/MISQ/2018/14108.
Yulin, Fang., Kai, H. Lim., Ying, Qian., Bo, Feng. (2018). System dynamics modeling for information systems research: Theory development and practical application, MIS Quarterly, 42(4), 1303-1329. DOI: 10.25300/MISQ/2018/12749.
Wetzels et al./Assessing Hierarchical Construct Models
USING PLS PATH MODELING FOR ASSESSING HIERARCHICAL CONSTRUCT MODELS: GUIDELINES AND EMPIRICAL ILLUSTRATION1
By: Martin Wetzels Department of Marketing Maastricht University Tongersestraat 53 6211 LM Maastricht THE NETHERLANDS [email protected]
Gaby Odekerken-Schröder Department of Marketing Maastricht University Tongersestraat 53 6211 LM Maastricht THE NETHERLANDS [email protected]
Claudia van Oppen SME Portal Maastricht University Tongersestraat 53 6211 LM Maastricht THE NETHERLANDS [email protected]
In this paper, the authors show that PLS path modeling can be used to assess a hierarchical construct model. They pro- vide guidelines outlining four key steps to construct a hier-
1Carol Saunders was the accepting senior editor for this paper.
archical construct model using PLS path modeling. This approach is illustrated empirically using a reflective, fourth- order latent variable model of online experiential value in the context of online book and CD retailing. Moreover, the guidelines for the use of PLS path modeling to estimate parameters in a hierarchical construct model are extended beyond the scope of the empirical illustration. The findings of the empirical illustration are used to discuss the use of covariance-based SEM versus PLS path modeling. The authors conclude with the limitations of their study and suggestions for future research.
Keywords: PLS path modeling, hierarchical construct model, empirical illustration, experiential value
Almost 25 years ago Noonan and Wold (1983, p. 283) observed: “Path analysis with hierarchically structured latent variables within the framework of PLS is at an early stage of development, and research is still under way.” Unfortunately, their observation is still a valid one, as applications and research into the use of hierarchical construct models using PLS path modeling are still limited. However, several authors have discussed both the theoretical and empirical contribu- tions hierarchical models can make (Edwards 2001; Edwards and Bagozzi 2000; Jarvis et al. 2003; Law and Wong 1999; MacKenzie et al. 2005; Petter et al. 2007), although almost exclusively in the realm of covariance-based structural equation modeling (SEM). Components-based SEM, or PLS path modeling, can also be used to estimate hierarchical con-
MIS Quarterly Vol. 33 No. 1, pp. 177-195/March 2009 177
Wetzels et al./Assessing Hierarchical Construct Models
struct models (Lohmöller 1989; Noonan and Wold 1983; Petter et al. 2007; Wold 1982). In this manuscript we will use PLS path modeling to construct a hierarchical construct model, show an empirical application, and provide guidelines for its use.2
For our empirical illustration we have chosen the construct of online experiential value, which has recently been advanced as a crucial driver of e-loyalty (Kim and Stoel 2004; Novak et al. 2000). When Mathwick et al. (2001, 2002) introduced, developed, and tested their experiential value scale, they referred to an experience-based value concept. Theoretically, their experiential value concept represents a fourth-order, reflective, hierarchical construct model that consists of intrinsic (hedonic) value and extrinsic (utilitarian) value as underlying dimensions (Babin et al. 1994; Holbrook and Hirschman 1982). Although the authors provide conceptual support for this hierarchical model, their empirical study only partially tests it.
Therefore, our main objective is to demonstrate that PLS path modeling can be used to estimate the parameters in a fourth- order, reflective, hierarchical construct model using online experiential value as an empirical illustration. This demon- stration extends the work by Mathwick et al. (2001) by specifying experiential value as a reflective, fourth-order latent variable with hedonic (intrinsic) and utilitarian (extrinsic) value as underlying dimensions at the third-order level. We include the resulting hierarchical model in a struc- tural model assessing its nomological validity. We use this application to demonstrate guidelines for assessing hier- archical models using PLS path modeling.
This paper is structured as follows. We first elaborate on the contributions of hierarchical construct modeling. In the next section we will discuss how hierarchical construct models can be estimated using structural equation modeling. Then we provide guidelines to build hierarchical construct models using PLS path modeling. Subsequently, we provide an empirical demonstration of the procedure suggested. We discuss how the guidelines suggested can be extended beyond
the scope of our empirical illustration. Moreover, we discuss the implications of our study by focusing on the conditions under which PLS path modeling might be more adequate than covariance-based SEM. Finally, we conclude with the limita- tions of the paper and suggestions for further research.
The Utility of Hierarchical Construct Models
Hierarchical constructs, or multidimensional constructs, as their discussion and application is often limited to a second- order hierarchical structure, can be defined as constructs involving more than one dimension (Edwards 2001, Jarvis et al. 2003; Law and Wong 1999; Law et al. 1998; MacKenzie et al. 2005; Netemeyer et al. 2003; Petter et al. 2007). As such, they can be distinguished from unidimensional con- structs, which are characterized by a single underlying dimen- sion (Netemeyer et al. 2003).
The utility of hierarchical construct models is based on a number of theoretical and empirical grounds (Edwards 2001). Proponents of the use of higher-order constructs have argued that they allow for more theoretical parsimony and reduce model complexity (Edwards 2001; Law et al. 1998; MacKenzie et al. 2005;). Edwards (2001) summarizes this argument as theoretical utility; theory requires general con- structs consisting of specific dimensions or facets. This is closely related to the trade-off between accuracy and generali- zation as suggested by Gorsuch (1983), who argues that “factors are concerned with narrow areas of generalization where the accuracy is great [whereas] higher-order factors reduce accuracy for an increase in the breadth of generali- zation” (p. 240). Law et al. (1998, p. 749) even state that “treating dimensions as a set of individual variables precludes any general conclusion between a multidimensional construct and other constructs.”
Moreover, hierarchical construct models allow matching the level of abstraction for predictor and criterion variables (Edwards 2001). Fischer (1980) refers to this as measure specificity, that is, predictor and criterion (latent) variables should be related to each other on the same level of abstrac- tion. For example, Chin and Gopal (1995) discuss three models, in which the intention to adopt GSS is explained by the belief in GSS adoption. Their first model links the underlying dimensions (relative advantage, ease of use, compatibility, and enjoyment) of belief in GSS adoption directly to intention to adopt GSS, without introducing a higher-order latent variable. The second model is a “molar” model, in which belief in GSS adoption is constructed as a
2 As far as terminology is concerned we will use the definition provided by
Bacharach (1989, p. 500) for constructs and variables: “a construct may be viewed as a broad mental configuration of a given phenomenon, while a variable may be viewed as an operational configuration derived from a construct.” Hence, we use the general term hierarchical construct model, while referring to manifest variables and latent variables for the operational configuration. Furthermore, we will use the terms reflective and formative to refer to causal relationship between latent variables and manifest variables (Edwards and Bagozzi 2000) as these are more prevalent in the information systems (Petter et al. 2007) and PLS literature (Chin 1998; Tenenhaus et al. 2005).
178 MIS Quarterly Vol. 33 No. 1/March 2009
Wetzels et al./Assessing Hierarchical Construct Models
latent variable with formative dimensions, while the third model is a “molecular” model, in which belief toward GSS adoption is constructed as a latent variable with reflective dimensions. Model 2 and Model 3 presented by Chin and Gopal show a fit regarding the level of abstraction, while Model 1 links dimensions directly to a potentially higher- order latent variable.
The conceptual grounds raised above are complemented by two empirical points: reliability and validity of measures of the multidimensional constructs (Edwards 2001). Typically, as the heterogeneity of the dimensions of the multidimen- sional construct increases, the internal consistency of the summed dimension scores will eventually be reduced. More- over, the construct validity of the dimension measures has been questioned, as it contains large amounts of specific and group variance, which are generally treated as error variance (see Law et al. 1998). Finally, proponents of higher-order constructs contend that such constructs exhibit a higher degree of criterion-related validity, especially if they serve as predictors.
Estimation of Hierarchical Construct Models
Edwards (2001) proposes an integrative analytical framework on the basis of (covariance-based) structural equation modeling (SEM), which allows for the simultaneous inclusion of higher-order (multidimensional) constructs and their dimensions as latent variables. In a structural model, the higher-order constructs may serve as either cause or effect by being embedded in a nomological network. This approach also allows us to derive the (indirect) effects of lower-order constructs, or dimensions, on outcomes of the higher-order construct as the pairwise product of loadings (or weights for formative constructs) and coefficients of the outcomes. Moreover, SEM allows for the explicit specification of the direction of the relationships between manifest variables and latent variables (Edwards and Bagozzi 2000).
Essentially, two models of higher-order (multidimensional) constructs can be distinguished on the basis of the directions of the relationship between manifest and latent variables (Law and Wong 1999):
(1) the factor model (Chin and Gopal: molecular model; Edwards: superordinate construct model; Jarvis et al.: principal factor model; Law et al.: latent model; MacKenzie et al.: common latent construct), and
(2) the composite model (Chin and Gopal: molar model; Edwards: aggregate construct; Jarvis et al.: composite latent variable model; Law et al.: aggregate model; MacKenzie et al: composite latent construct model).
For the factor model, or reflective construct model, the manifest variables are affected by the latent variable(s) (LVj→MVi), whereas for the composite model, or the formative construct model, this relationship is reversed (LVj←MVi).
For the reflective construct model, higher-order, or hier- archical, latent variable models can be specified as an alter- native to group-factor models, i.e., a latent variable model for which all first-order latent variables are correlated, or a first- order confirmatory factor analysis (Bollen 1989; Guinot et al. 2001; Hunter and Gerbing 1982; Marsh and Hocevar 1985; Rindskopf and Rose 1988). Basically, a hierarchical model imposes an alternative structure on the pattern of correlations (covariances) among lower-order latent variables (or factors) of the group-factor model. As such, the higher-order model represents a restriction of the group-factor model, which allows for correlation of the lower-order latent variables (or factors; Rindskopf and Rose 1988). For example, a second- order model can be specified by the following two equations:
(1) yi = Λy A ηj + gi
(2) ηj = Γ A ξk + ζj
The first equation defines the manifest variables