You are on page 1of 12

ARTICLE IN PRESS

International Journal of Information Management 28 (2008) 102113 www.elsevier.com/locate/ijinfomgt

Customer Knowledge Management and E-commerce: The role of customer perceived risk
Molina-Castillob Carolina Lopez-Nicolasa,, Francisco Jose
b

n de Empresas y Finanzas, Universidad de Murcia, Campus de Espinardo, 30100 Murcia, Spain Departamento de Organizacio n e Investigacio n de Mercados, Universidad de Murcia, Campus de Espinardo, 30100 Murcia, Spain Departamento de Comercializacio

Abstract The present research is designed to gain a deeper understanding of Customer Knowledge Management (CKM) tools inside the e-commerce context. The relationship between the CKM literature and the e-commerce literature is evaluated through several user characteristics such as risk preference, Internet preference and Internet knowledge and their impact on customers online perceived risk and purchase intentions depending on the presence of certain CKM tools on the web site. The empirical study is based on a survey of 276 customers with previous online experience. By using multidimensional analysis, this study shows that the customers perceived risk associated with different CKM tools plays an important role in explaining certain customer online behaviour. Therefore, the implications of CKM tools for e-commerce activity are demonstrated and the managerial implications are highlighted. r 2007 Elsevier Ltd. All rights reserved.
Keywords: CKM tools; E-commerce; Perceived risk; Purchase intention; Customer perceptions

1. Introduction In modern organizations, knowledge is the fundamental basis of competition (Zack, 1999), and information technology (IT) is a necessity (Bose, 2000) critical for managing knowledge (Ofek & Sarvary, 2001). In the new context, two major factors determine the future survival or success of organisations: electronic commerce (Gupta, Su, & Walter, 2004) and the knowledge from customers (Tsai & Shih, 2004), encouraging the adoption of e-commerce and the use of the Internet as a platform to access and collect important knowledge from customers. In other words, the success of e-commerce increasingly depends on knowledge management (Borges, Almeida, Gomes, & Cabral, 2007; Saeed, Grover, & Hwang, 2005). Customer Knowledge Management (CKM) is the application of knowledge management (KM) instruments and techniques to support the exchange of knowledge between
Corresponding author. Tel.: +34 968 363762; fax: +34 968 367537.

E-mail addresses: carlopez@um.es (C. Lopez-Nicolas), fjmolina@um.es (F.J. Molina-Castillo). 0268-4012/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.ijinfomgt.2007.09.001

an enterprise and its customers (Kolbe & Geib, 2005; Rollins & Halinen, 2005; Rowley, 2002), enabling the company to make appropriate strategic business decisions (Rowley, 2002; Su, Chen, & Sha, 2006). However, there is still a need to further elaborate on the concepts of customer knowledge and CKM (Rollins & Halinen, 2005), since the critical role of KM in gaining competitive advantage in the market (Ofek & Sarvary, 2001) and within the e-commerce context (Du Plessis & Boon, 2004; Tsai & Shih, 2004) is far from fully understood. Knowledge, dened as information combined with experience, context, interpretation and reection (Davenport, De Long, & Beers, 1998), can be divided into explicit knowledge and tacit knowledge (Nonaka, 1994). Specically, customer knowledge can also be classied as knowledge for, about or from the customer (Maswera, Dawson, & Edwards, 2006; Salomann, Dous, Kolbe, & Brenner, 2005; Su et al., 2006). KM is the explicit and systematic management of vital knowledge and its associated processes of creation, organisation, diffusion, use and exploitation (Skyrme, 2001) and CKM is the external perspective of KM (Rollins & Halinen, 2005). In order to

ARTICLE IN PRESS
C. Lopez-Nicolas, F.J. Molina-Castillo / International Journal of Information Management 28 (2008) 102113 103

put KM and CKM into practice, some organisations may implement initiatives related to more humanistic practices, while others are based on IT that may be hosted in the corporate Intranet and/or web site (Wang, 2001). Our research focuses on the latter, where knowledge ows into and out of the company through certain CKM tools hosted on the rms web site (Shared databases, Document repositories, Workow applications and Discussion forums) but whose implications for customer perceptions need greater clarication for managerial purposes. Executives should use KM and e-commerce principles to complement each other, as a way of electronic CKM, making it possible to obtain priceless information and knowledge from customers about their needs and purchase intentions. Embedding KM programs that customers may access within a companys web site may actually be an obstacle to the increase of e-commerce (Bose, 2000), and this suggests the need for more research in this area. Therefore, by adopting an external KM perspective (CKM), the aim of our investigation is to assist organisations in their Web initiatives for managing customer knowledge. Online shopping is developing rapidly today and e-commerce initiatives have been found to increase the value of the rm. Researchers, however, agree that in fact the amount of money involved remains very low (Cases, 2002; Gupta et al., 2004). The perceived risk of conducting transactions online has recently been considered to be the most important factor in explaining consumers reluctance to complete simple online purchase transactions (Forsythe & Shi, 2003). In this sense, we are concerned with the fact that perceived risk in different CKM Web tools may inuence the success of e-commerce projects in terms of the purchase intentions of consumers. Thus, our thesis is that hosting certain CKM tools on the corporate web site, such as Shared databases, Document repositories, Workow applications and Discussion forums, could cause an increase in perceived Web risk and, in turn, a backward step in customers purchase intentions through that site. We also aim at analysing the role of other variables, such as a customers risk preference, Internet knowledge and Internet preference, on the model. The paper is organized as follows. First, the most common Web tools used in CKM are reviewed, considering the potential differences between CKM tools in terms of perceived risk (Section 2). Next, relationships between perceived risk associated to each CKM tool, purchase intention linked to each CKM application and users characteristics, namely, their risk preference, Internet knowledge and Internet preference, are discussed, proposing a theoretical model to be empirically tested (Section 3). Then, the methodology and the measures used in the survey are explained (Section 4) and research ndings are shown (Section 5). Finally, conclusions and limitations are summarised and future research lines presented (Section 6).

2. Differences between CKM tools in e-commerce 2.1. Online CKM tools KM is especially being adopted by companies who have invested in the Internet (Borges et al., 2007), in order to manage customer knowledge. Through CKM Web applications, organisations may obtain vital knowledge, adding an extra dimension to marketing research activity (Cheung & Huang, 2002) and improving customer service. Managing the collection, storage and distribution of relevant knowledge requires the integration of KM and CRM resulting in CKM (Kolbe & Geib, 2005). Web-based customer data become an important source for KM (Chou & Lin, 2002) and the challenge is to convert customer data and information to knowledge (Martin, 2001; Maswera et al., 2006; Rowley, 2002), in order to segment the market (Davenport, Harris, & Kohli, 2001; Su et al., 2006), to customize products and marketing (Martin, 2001; Maswera et al., 2006), to provide exceptional customer service (Martin, 2001; Shah & Murtaza, 2005), to shorten product development cycles and reduce the risk of the DNP process (Su et al., 2006), to impact the customers perception of service quality (Saloman et al., 2005) and achieve greater customer loyalty and retention (Martin, 2001). There are many solutions to managing both explicit and tacit customer knowledge (Davenport et al., 2001). Recent literature (Maswera et al., 2006; Romano & Fjermestad, 2003; Shah & Murtaza, 2005) suggests that the most common Web tools in companies CKM efforts are Shared databases, Document repositories, Workow applications and Discussion forums.

Shared databases: Businesses want its partners and customers to be able to view and update databases (Shah & Murtaza, 2005). For example, Cisco Systems provides its customers access to the same internal database that is used by its employees (Saeed et al., 2005). Shared databases are considered to be important tools of the trade for anyone in the supply chain (Saeed et al., 2005). Document repositories: Also called knowledge repositories, they typically store documents with knowledge embedded in them (Kwan & Balasubramanian, 2003) and may also be accessed via rms web sites so that external agents can gain access to important catalogues, manuals and documents to make buying decisions. The objective is to externalise knowledge, store it in repositories and make it explicit and accessible, for later and broader access, across the organisation via the corporate intranet (Kwan & Balasubramanian, 2003), as an example of a codication strategy for managing knowledge (Hansen, Nohria, & Tierney, 1999). For instance, Benetton provides web users (through www.benetton.com) with important documents, such as their product catalogue,

ARTICLE IN PRESS
104 C. Lopez-Nicolas, F.J. Molina-Castillo / International Journal of Information Management 28 (2008) 102113

photo gallery, videos showing its infrastructure, or its corporate social responsibility policy.

Table 1 Dimensions of perceived risk Dimension Denition

Workow applications: These may dened as the automation of a business process, in whole or part, during which documents, information or tasks are passed from one participant to another for action, according to a set of procedural rules (Workow Management Coalition, 1999). Some companies are beginning to notify customers, by email or SMS, when the product or service provided gets to the next step in the production and delivery processes. For example, UPS offers parcel tracking services through www.ups.com and Dell computers sends emails to customers about new product development phases in order to let them know the situation of the product before it is received. Discussion forums: Web discussion forums permit the participation of a larger and more diverse set of people and information resources (DeSanctis, Fayard, Roach, & Jiang, 2003), thus allowing them to express their needs, doubts and purchase intentions (Maswera et al., 2006) and helping specialised knowledge workers to make sense of other community perspectives (Hayes & Walsham, 2001) and to develop new products and services. Customers provide information and tacit knowledge about themselves during engagement in an online community (Rowley, 2002) and companies can monitor online chat to make the site more relevant for their customers (Ofek & Sarvary, 2001; Rowley, 2002). For instance, Transcend, offer the opportunity to customer to include questions on the Discussion forums and propose new alternative to data storage not only on product functionality but also on product design.

Associated with the product Technical risk The probability that a purchased product results in failure to function as expected Service risk The probability that the rm will not offer a good service in the future Social risk The probability that a product purchased results in the disapproval of family or friends Psychological risk The probability that a product results in inconsistency with self-image Associated with the place Performance risk Financial risk Time risk Delivery risk The probability that the buying process does not perform as expected The probability that a purchase results in loss of money or other resources The probability that a purchase results in loss of time to buy or retain the product The probability that a purchase results in problems when delivering the product to the customer

In conclusion, the volume of qualitative data available via corporate web sites is growing and rms are looking forward to extracting and understanding users thought processes, wants, needs, and purchase intentions (Romano & Fjermestad, 2003) contained in those CKM Web applications. Nonetheless, including certain KM tools in corporate web sites could be affecting other variables such as customers perceived risk or customers purchase intentions and, in the long run, the rms sales. 2.2. Perceived risk in CKM tools One of the main concerns expressed in the academic literature is related to the risk perceived by customers when buying a specic good, both in traditional shopping and in online environments. Consumer behaviour involves risk since any action of a consumer will produce consequences that he or she views with some amount of uncertainty (Bauer, 1960). In this sense, perceived risk involves the amount that would be lost if consequences of an act were not favourable, combined with individuals subjective feeling of the likelihood that the consequences will actually

be unfavourable (Mitchell, 2001). There is a consensus in the literature that there are different dimensions comprising the perceived risk construct (Table 1). Basically, risk can be associated with the product and risk associated with the place where the product is offered, and in e-commerce the retail channel is the Internet. Comparing perceived risk in traditional shopping to new online environments, the risk level associated with certain dimensions might be increased, while other risk forms may appear only in the online context (Forsythe & Shi, 2003). Specially signicant in Internet shopping is the risk associated with the product and security (Doolin, Dillon, Thompson, & Corner, 2005; Pavlou, 2003), due to three elements which characterise this context: a remote source (namely the site on which the transaction takes place), an interactive medium for sending the message and an online command mode (Cases, 2002). However, online applications, such as the CKM tools examined here, may be good (or bad) risk-relievers and Doolin et al. (2005) recommend Internet retailing web sites to include certain features that reduce the perceived risk. For instance, a Discussion forum hosted on the corporate web site allows users to exchange comments, recommendations and word of mouth about the product, the company and the site, and are thus an important mechanism to reduce consumers perceived risk (Garbarino & Strahilevitz, 2004). Also, the presence of electronic repositories containing product information and demos on a web site may reduce the product risk perceived by the user (Cases, 2002), while access to online documents where security and privacy policies are clearly disclosed might mitigate consumers perceived privacy risk (Doolin et al., 2005), the most signicant perceived risk dimension in online shopping. In contrast, hosting other CKM tools on a web site may augment the complexity of the site (Chen &

ARTICLE IN PRESS
C. Lopez-Nicolas, F.J. Molina-Castillo / International Journal of Information Management 28 (2008) 102113 105

Macredie, 2005) and increase the risk perceptions of users. This may be the case with Shared databases, where authorised users and, unfortunately, hackers may have access to important information and knowledge about customers, thus making it possible to offer condential data to all internauts without their knowing. In this situation, users will consider the web site to be less safe (Conchar, Zinkhan, Peters, & Olavarrieta, 2004) and perceive a higher risk on the web site that hosts Shared databases. Finally, Workow applications on a web site automate specic processes, some containing consumers private information which may be accessed by unauthorised people. In this situation, users may perceive a higher level of risk when using that web site with Workow tools. For this reason, due to the specic characteristics of every CKM tool described before, we posit that there could be a distinct risk level associated with each one when hosted on a corporate web site. H1. The customers perceived risk associated to each CKM tool hosted on a web site will be different. 3. Implications of CKM tools in e-commerce The Internet is profoundly changing KM, promoting it from a trend to an e-business reality (Borges et al., 2007). The recent literature considers the Internet to be a new retail channel (Gupta et al., 2004), with great potential for commercial usage (Cheung & Huang, 2002). However, most online consumers use information gathered online to make purchases off-line (Forsythe & Shi, 2003; Shim, Eastlick, Lotz, & Warrington, 2001), which means that the amount of money involved in e-commerce remains very low (Saeed et al., 2005). Many factors may explain why Internet browsers do not become online shoppers, but the present article focuses on perceived risk and users characteristics in order to shed light on the variables affecting consumers purchase intentions in the online context (Fig. 1). 3.1. Perceived risk Among the reasons commonly cited for consumers aborting purchase attempts are a reluctance to supply

personal and credit card information, technical problems with web sites, and problems in locating products (Shim et al., 2001). Consumers perceptions of risk are considered to be central to different steps in the buying process: their evaluations, choices, and behaviours (Garbarino & Strahilevitz, 2004), since consumers are more often motivated to avoid mistakes than to maximise utility in purchasing (Conchar et al., 2004). Thus, in online contexts, an increase in the risk perceived by customers could reduce their intention to buy through that web site. Perceived risk toward a product category has been shown to be negatively associated with purchase intentions toward that product category (Westland, 2002). Similar logic should hold true for perceived risk toward a particular shopping channel. Indeed, several studies have suggested that risk perceptions toward remote purchasing methods can affect related shopping behaviour (Mitchell, 2001). Thus, consumers who perceive fewer risks or concerns toward online shopping are expected to make more online purchases than more risk-laden consumers (Miyazaki & Fernandez, 2001). The perceived risk associated with online transactions may reduce perceptions of behavioural and environmental control, affecting negatively transaction intentions (Forsythe & Shi, 2003). Perceived risk has been found to have a negative inuence on consumers attitudes or intentions to purchase online (Novak, Hoffman, & Yung, 2000). Given the uncertain context of e-commerce, it is expected that perceived risk would lower consumers intentions to use Internet sites for transactions (Pavlou, 2003). The thesis of the present research is that hosting CKM tools such as Shared databases, Document repositories, Workow applications and Discussion forums in a web site could cause an increase in perceived Web risk and, in turn, reduce costumers purchase intentions on that site. These statements give us the chance to formulate the following hypothesis: H2. The higher the customers perceived risk associated with a CKM tool hosted in a web site, the lower the purchase intention from that customer.

Risk preference

H3 Perceived Risk associated to each CKM tool H1

H4 Internet knowledge H6

H2 Purchase intention H5

Internet preference

Fig. 1. CKM tools in e-commerce.

ARTICLE IN PRESS
106 C. Lopez-Nicolas, F.J. Molina-Castillo / International Journal of Information Management 28 (2008) 102113

3.2. User characteristics 3.2.1. Risk preference As Conchar et al. (2004) explain in their exhaustive review of perceived risk, risk preference has also been studied, for instance, as risk tolerance (Sitkin & Pablo, 1992) or risk propensity (Forsythe & Shi, 2003). Risk preference is a psychological feature of a users personality and may be dened as a decision-makers tendency to take (or avoid) risks (Conchar et al., 2004). Regarding the online environment, Chen and He (2003) empirically found a similar link between risk preference and risk perceptions. Basing their study on structural equation modelling, they concluded that the higher a persons risk preference, the lower his/her perceived risk. Nevertheless, decision-makers who enjoy the challenge that risks entail will be more likely to undertake risky actions (Sitkin & Pablo, 1992), meaning that risk preferring individuals will be willing to incur high risk and will complete transactions on the most risky orders (Westland, 2002). In line with this, Conchar et al. (2004) state that a person with high-risk afnity will prefer an alternative perceived as more risky. In those situations, users who are risk-seekers will perceive higher levels of risk than risk-averse individuals. Thus, we may hypothesize a positive relationship between risk preference and perceived risk. H3. The higher the users risk preference, the higher the perceived risk associated with a CKM tool hosted in a web site. 3.2.2. Internet knowledge Often called Internet experience, this is dened as the consumers skill or ability obtained by visiting several web sites and using various value-added services offered on a broad range of web sites, and not as experience with one particular web site (Nysveen & Pedersen, 2004). Consumers knowledge about the Internet is important in understanding customers perceptions, attitudes, and behaviour in online environments (Shim et al., 2001). Specically, Internet experience contributes to more effective use of web site applications in a way that experienced Internet users have more positive attitudes to using a web site (Chen & Macredie, 2005). Many marketers believe that experience gained through simple usage of the Internet for non-purchase purposes such as information gathering and non-commercial communication will lead consumers to discover that privacy and security risks are often exaggerated (Miyazaki & Fernandez, 2001). It has been found that the more frequently a consumer uses the Internet, the more knowledgable he/she has in using the Internet and the consumer feels less risk associated with the Internet (Chen & He, 2003). Based on previous research, we posit that Internet knowledge may be a factor in reducing users risk perceptions in the online context.

H4. The higher the users Internet knowledge, the lower the perceived risk associated with a CKM tool hosted in a web site. 3.2.3. Internet preference Humancomputer experiences are usually playful and exploratory (Bierly & Daly, 2002), expanding the time and effort devoted to exploring new options and experimenting with new possibilities. In this sense, the Web may be characterized as pleasurable, fun, enjoyable and as something that enables the Web user to escape from reality (Chung, Chen, & Nunamaker, 2005). Internet preference relates to the users personality feature associated with enjoying with Internet exploration and surng. This exploratory behaviour positively inuences the users attitudes toward the web site (Das, Echambadi, McCardle, & Luckett, 2003) and, in turn, may be a signicant factor in e-commerce acceptance and online purchase intentions (Richard & Chandra, 2005). H5. The higher the users Internet preference, the higher the purchase intention from the user. On the other hand, Internet preference may be a consequence of the users Internet knowledge and experience. As consumer knows more about this channel, he/she enjoys more when navigating on the Internet (Cheung & Huang, 2002). It has been found recently that people skilled at using the Internet really enjoy exploring web sites they hear about, thus showing a higher Internet preference and, indirectly, improving attitudes towards the site (Chen & He, 2003). That is, Internet skills have a positive inuence on exploratory behaviour (Richard & Chandra, 2005). Also, Das et al. (2003) found empirically that users considered as experts or experienced in navigating the Web did use the Web for fun and excitement, as a recreational way to relax and to spend their time. Thus, based on the literature, we may hypothesize that Internet knowledge and experience may positively inuence Internet preference. H6. The higher the users Internet knowledge, the higher the users Internet preference. All the links hypothesized basing on the literature review are shown in the model graphically presented in Fig. 1. The theoretical framework we propose integrates KM and e-commerce areas by considering the impact CKM tools may have on different key variables of e-commerce. 4. Methodology 4.1. Sample and data collection In order to contrast our hypothesis we conducted an experiment among Internet customers. A sample of 276 undergraduate students from different courses at a large university was chosen. The sample was selected with an attempt to concentrate on future business leaders who are

ARTICLE IN PRESS
C. Lopez-Nicolas, F.J. Molina-Castillo / International Journal of Information Management 28 (2008) 102113 107

familiar with the kind of instruments used in this research and who, nowadays, play an important role as Internet customers. A self-administered questionnaire was prepared for use in the survey, and this was pre-tested on 10 IT and Business experts. A number of suggestions were obtained on how to improve the questionnaire substantially. Once the modications were included, the questionnaire was given to the students. The survey instrument started with several questions concerning previous e-commerce experience, and these were followed by sections where each student was asked to value the perceived risk and purchase intentions associated with each CKM tool. Students were provided with a description of each CKM tool (Shared databases, Document repositories, Workow applications and Discussion forums), the place where they usually appear on a web site and their main utility for rms and customers. Finally, customers were rewarded with entry into a contest for a DVD player in order to increase their involvement with the research project. 4.2. Measure development and scale properties The variables for this research were measured using multi-item scales tested in previous studies. The response categories for each scale were ranked between 0 (strongly disagree) and 10 (strongly agree) because pre-testing showed that items were better understood when valuing each of the concepts from 0 to 10, since Spanish students are normally marked in their courses using a similar range. This procedure has also been consistently applied in the literature. For measuring the perceived risk component we drew upon the work of the rst author that proposed this construct, Bauer (1960), together with other articles which, in recent years, have also paid attention to it, namely, Mitchell (2001). Finally, for perceived risk measurement, we considered four of the components most frequently cited in the literature and related to the risk associated with the place that offers the product (performance risk,

nancial risk, time risk, delivery risk). Risk preference was assessed through the work of Chen and He (2003). Internet knowledge was measured through the work of Novak et al. (2000) and Internet preference based on the work of McKnight, Choudhury, and Kacmar (2002). Purchase intention was measured through two items based on the study of Chen and He (2003). A detailed description of the scales can be found in the Appendix. Before testing the hypotheses, we discuss the scale reliability of all the measures in this study (Table 2). We conducted a conrmatory factor analysis (CFA) including the independent and dependent constructs with Lisrel 8.5 for the Shared database model (w2(94) 182.49, CFI 0.96, IFI 0.96, NFI 0.92, NNFI 0.95, GFI 0.92, RMSEA 0.059, RMR 0.052), the Document repositories model (w2(94) 141.04, CFI 0.98, IFI 0.98, NFI 0.94, NNFI 0.97, GFI 0.94, RMSEA 0.043, RMR 0.045), the Workow model (w2(94) 181.91, CFI 0.96, IFI 0.96, NFI 0.92, NNFI 0.94, GFI 0.92, RMSEA 0.058, RMR 0.044), and the Discussion forum model (w2(94) 173.39, CFI 0.96, IFI 0.96, NFI 0.92, NNFI 0.95, GFI 0.93, RMSEA 0.055, RMR 0.050). The principal adjustment indices (absolute, incremental and parsimony) of the vefactor model for each CKM tool suggest a good t of the specication for our measures of the independent and dependent variables. All of the loadings for the items on their respective constructs were large and signicant (smallest t-value 3.62), which provides evidence of convergent validity (Bagozzi & Yi, 1988). Regarding the nature of the individual parameters and the internal structure of the model, all factor loadings were signicant and all of them exceeded the 0.7 level required as a basis for research. The reliability of the multi-item scales was assured by calculating the composite reliability index suggested by Bagozzi and Yi (1988) and with the average variance extracted index proposed by Fornell and Larcker (1981). As shown in Table 2, both indexes are inside the recommendations of the

Table 2 Descriptive statistics and reliability Mean S.D. No. of items Cronbachs remain alpha 4 4 3 4 4 4 4 2 2 2 2 0.90 0.80 0.87 0.78 0.81 0.81 0.80 0.83 0.76 0.78 0.80 Eigenvalue Lowest t-value 14.56 9.98 14.11 8.80 11.38 9.16 10.36 5.81 6.67 7.04 3.62 SCRa AVEb

Internet knowledge Risk preference Internet preference Perceived risk of Shared databases Perceived risk of Document repositories Perceived risk of Workow Perceived risk of Discussion forums Purchase intention of Shared databases Purchase intention of Document repositories Purchase intention of Workow Purchase intention of Discussion forums
a b

6.11 4.78 5.63 4.11 3.24 3.63 3.08 5.01 5.57 5.87 5.79

2.01 2.19 2.51 2.21 1.95 2.19 1.98 2.37 2.25 2.34 2.44

3.10 2.15 2.45 2.41 2.59 2.53 2.49 1.72 1.61 1.63 1.67

0.91 0.81 0.89 0.80 0.82 0.82 0.80 0.84 0.78 0.79 0.80

0.72 0.60 0.73 0.50 0.53 0.53 0.50 0.63 0.64 0.65 0.63

Scale composite reliability. Average variance extracted.

ARTICLE IN PRESS
108 C. Lopez-Nicolas, F.J. Molina-Castillo / International Journal of Information Management 28 (2008) 102113

literature, which provides evidence of a good adjustment of each construct. In addition, Evidence of discriminant validity among the dimensions of each construct was provided by three different procedures recommended in the literature as follows: (1) when a 95% condence interval constructed around the correlation estimate between two latent variables never includes the value 1 (Anderson & Gerbing, 1988); (2) when the hypothesised four-factor model has a signicantly better t to the data than an alternative model in which the correlation estimate between latent constructs is constrained to the value 1 (Anderson & Gerbing, 1988); (3) when the individual average variance extracted for each latent variable exceeds the squared correlation between both latent variables (Fornell & Larcker, 1981). 5. Research ndings 5.1. Differences between CKM tools in e-commerce To test whether or not differences exist between the variables for each CKM Web tool, a statistical analysis based on the mean differences among the constructs was conducted. Results revealed that difference exists in perceived risk among all the CKM tools considered except between Document repositories and Workow. This supports hypothesis H1 about the distinct perceived risk associated to each CKM tool. Moreover, as can be seen in Fig. 2, the higher customer perceived risk appears when Shared database tools are hosted in a web site. On the contrary, Discussion forum tools produce a lower perceived risk on the part of customers. This means that the presence of Discussion forums, where word of mouth can be shared, may relieve the risk perceived online. This nding is similar to that described in Cases (2002), who proposed that sharing word of mouth online is a
4.5 4 3.5 3 2.5 2 1.5 1 0.5 0 4,11

risk-reliever. In addition, the empirical data support the idea, previously discussed in the literature review, that the fact that web sites use Shared databases and/or Workow applications means that they are perceived by users as riskier than other web sites where Document repositories and/or Discussion forums are hosted. 5.2. Implications of hosting CKM tools in e-commerce The proposed structural model for each CKM tool is specied from the hypothesized relationships in Fig. 1, discussed in the text as H2H6. Conventional maximum likelihood estimation techniques were used to test the model. However, it is generally agreed that researchers should compare rival models and not just test the performance of a proposed model (Bagozzi & Yi, 1988). Our proposed ve-factor model was compared with another model that also estimates the relation of Internet knowledge with purchase intention. The underlying assumption built into this alternative model is based on the proposal of several authors (e.g. Miyazaki & Fernandez, 2001) that the higher the Internet knowledge, the higher the probability of shopping online. Therefore, we test our theoretical model (TM) against an alternative model specication (AM) that considers this extra relationship. Anderson and Gerbing (1988) recommend this procedure and suggest the use of a Chi-square difference test (CDT) to test the null hypothesis: TMAM 0. Compared with a less parsimonious model (AM) that also considers the direct relationship between Internet knowledge and purchase intentions, a non-signicant CDT would lead to the acceptance of the more parsimonious TM. The non-signicant change in Chi-square between our model (TM) and the alternative one (AM) for every CKM tool, leads us to consider TM as a better specication.
CKM tool 3,08

3,61 3,23

Shared databases

Documents Repositories

Workflow

Discussion Forums

Mean values are expressed for each CKM tool Shared databases and documents repositories Shared databases and workflow Shared databases and discussion forums Document repositories and workflow Document repositories and discussion forums Workflow and discussion forums Significance levels: ***p<0.01 **p<0.05 *p<0.10 Mean differences 0.86*** 0.49*** 0.99*** 0.38*** 0.14 0.52*** T-student 7.24 3.73 7.14 3.64 1.53 4.77

Fig. 2. Perceived risk for each CKM tool.

ARTICLE IN PRESS
C. Lopez-Nicolas, F.J. Molina-Castillo / International Journal of Information Management 28 (2008) 102113 109

Results indicate that the t of our proposed model was much better than the t of the respecied model for every CKM tool. According to this, the t of the model for Shared databases is satisfactory (w2(98) 185.23, CFI 0.96, IFI 0.96, NFI 0.92, NNFI 0.95, GFI 0.92, RMSEA 0.05, RMR 0.05) and all the hypothesis were conrmed, thus revealing the mediating role of perceived risk and Internet preference in our model. Moreover, an indirect effect was found between Internet knowledge and purchase intention (0.10; po0.05), thus demonstrating the positive effect of this variable on the probability of the customer of buying online. The overall adjustment for the Document repositories also offer a good t (w2(98) 153.48, CFI 0.97, IFI 0.97, NFI 0.93, NNFI 0.97, GFI 0.93, RMSEA 0.04, RMR 0.05) and, similarly to what happened with the previous model, all the hypotheses were conrmed. Similarly, an indirect effect between Internet knowledge and purchase intention was also found (0.09; po0.05). In contrast to the previous model, the Workow model offered different ndings. The overall t of the model was acceptable (w2(98) 194.63, CFI 0.95, IFI 0.95, NFI 0.91, NNFI 0.94, GFI 0.92, RMSEA 0.06, RMR 0.05), but in terms of the hypotheses, our results conrm that the relationship between Internet knowledge and perceived risk (g12 0.08, p40.10) and between Internet preference and purchase intention (b32 0.02, p40.10) were not supported. This means that an adequate Internet knowledge does not necessarily lead to a reduction in the perceived risk associated to Workow tools. Another important nding is that Internet knowledge does not relate to customer purchase either directly or indirectly. Finally, the model for Discussion forums also offers unexpected results. Even though the overall t of the structural model is inside the recommendations of the literature (w2(98) 178.56, CFI 0.96, IFI 0.96,

NFI 0.92, NNFI 0.95, GFI 0.92, RMSEA 0.05, RMR 0.05) one of the hypotheses was not conrmed; specically, the one that relates the perceived risk associated to Discussion forums tools and purchase intention. This means that managers should not be discouraged from including this type of CKM tool, because they do not only lead to lower purchase intention based on higher levels of perceived risk, but there is also a second effect where purchase intention increases when customers have a preference for the Internet. On the other hand, when comparing the four models (one model for each CKM tool) shown in Figs. 36, some interesting results are found. First, the impact of Internet knowledge on Internet preference is similar in every CKM tool, with the estimated coefcient for this link being around 0.5 in all of the models. So, we may state that Internet knowledge is a good predictor for Internet preference for any CKM tool. Second, the strength of the impact that risk preference has on perceived risk is different depending on the CKM tool considered. Specically, estimated coefcients are higher in the case of Discussion forums and Document repositories rather than in the case of Shared databases and Workow applications. So, the link between risk preference and perceived risk is stronger when the web site offers Discussion forums and Document repositories and weaker when the company provides online access to Shared databases and Workow applications. Finally, the results show that the inclusion of Discussion forums is unique among CKM tools in not having an impact negatively on customers purchase intentions. This nding, together with the fact that this CKM tool has been proven to be a risk-reliever, makes Discussion forums the most advisable CKM Web application. 6. Conclusions and managerial implications Many organisations consider KM to be the fundamental basis of competition (Zack, 1999) and a critical enabler of

Risk preference

11 = 0.12* Perceived Risk shared databases

Significant path Not significant path Alternative path 31=-0.11** Purchase intention shared databases

12=-0.11* Internet knowledge

22=0.52***

Internet preference 2(98) = 185.23

32=0.14**

CFI=0.96 IFI=0.96 NFI=0.92 NNFI=0.95 GFI=0.92 RMSEA=0.05 RMR=0.05 Significance levels: ***p<.01 **p<.05 *p<.10
Fig. 3. Shared databases tools in e-commerce.

ARTICLE IN PRESS
110 C. Lopez-Nicolas, F.J. Molina-Castillo / International Journal of Information Management 28 (2008) 102113

Risk preference

11=0.25*** Perceived Risk document repositories

Significant path Not significant path Alternative path 31=-0.16** Purchase intention Document repositories Internet preference 2(98) = 153.48 32=0.11*

12=-0.15** Internet knowledge

22=0.53***

CFI=0.97 IFI=0.97 NFI=0.93 NNFI=0.97 GFI=0.93 RMSEA=0.04 RMR=0.05 Significance levels: ***p<.01 **p<.05 *p<.10
Fig. 4. Document repositories tools in e-commerce.

Significant path Risk preference 11=0.12* Perceived Risk workflow Not significant path Alternative path 12=-0.08 Internet knowledge 32=0.02 31=-0.20** Purchase intention workflow 22=0.52*** Internet preference

2(98) = 194.63 CFI=0.95 IFI=0.95 NFI=0.91 NNFI=0.94 GFI=0.92 RMSEA=0.06 RMR=0.05 Significance levels: ***p<.01 **p<.05 *p<.10
Fig. 5. Workow tools in e-commerce.

Significant path Risk preference 11=0.25** Perceived Risk discussion forums Not significant path Alternative path 12=-0.15** Internet knowledge 32=0.13* 31=-0.09 Purchase intention Discussion forums 22=0.53*** Internet preference

2(98) = 178.56 CFI=0.96 IFI=0.96 NFI=0.92 NNFI=0.95 GFI=0.92 RMSEA=0.05 RMR=0.05 Significance levels: ***p<.01 **p<.05 *p<.10
Fig. 6. Discussion forums in e-commerce.

ARTICLE IN PRESS
C. Lopez-Nicolas, F.J. Molina-Castillo / International Journal of Information Management 28 (2008) 102113 111

good customer support and service (Shah & Murtaza, 2005). Besides, KM capabilities rely strongly on IT infrastructure in order to improve customer response and provide faster decision-making (Chung et al., 2005). Smart companies seek knowledge about and from their customers (Chen & Macredie, 2005) and web sites can be the rst point of contact between a company and its customers (Chou & Lin, 2002) and the means to obtain knowledge about and from customers (Maswera et al., 2006). By integrating KM into their e-commerce activities, as a way of online CKM, rms can automate existing processes and dramatically reduce cycle times throughout the supply chain; they can enhance communication, collaboration, and corporation between knowledge teams (including virtual teams) using intranet technologies and between the organisation and members of its external constituent organisations using extranet technologies. From present research, it may be drawn the conclusion that incorporating certain web site features has a positive impact on customer perceptions, as suggested by Heinze and Hu (2006), and that the implementation of Internet based CKM will positively impact on e-business performance, as Borges et al. (2007) have recently found, in terms of online purchase intentions. Nevertheless, we have demonstrated that certain CKM tools may be harmful for the organisation, as examined more fully below. The results of this research are essential for academic and managerial purposes because they try to ll, to some extent, the gap that exists between KM and e-commerce activity, by analysing the antecedents and consequences of CKM Web tools hosted in corporate web sites. Moreover, this research extends the literature that identies a negative relation between perceived risk associated with certain CKM tools and purchase intentions in the online context. On the other hand, managers should take into account the implications of hosting some CKM applications on their web sites, because there could be an important effect on customer perception of a web site or on the nal sales volume. The empirical ndings show that there is an important link between KM and e-commerce, especially regarding the differences between CKM tools hosted on a web site, in terms of customers perceived risk. Moreover, results reveal that hosting Document repositories or Discussion forum tools in the corporate web site constitutes a signicant risk-reliever, in comparison to Shared databases and Workow tools. We have also found that the impact of customers perceived risk on purchase intention is not the same for every CKM tool considered. Specically, results suggest that hosting Discussion forums enhances the probability of customer purchases in contrast with the situation of Shared databases, Document repositories or Workow applications. Consequently, Discussion forums have been found as the CKM tools that are most commendable in e-commerce initiatives. Despite its important contributions for academics and practitioners, this study also has some limitations. We conducted our study with 276 students, so there are some

problems with the external validity of the results. For that reason, it could be interesting to test this research with other customers in order to generalize our ndings. Moreover, it will be useful to replicate this study using an online survey and, if possible, with data from real rms selling products on the Internet. Also, examining differences between sectors or types of products may be helpful for managerial implications. Finally, further research is needed about how some other variables, such as demonstrations and guarantees (Gupta et al., 2004) or user gender (Garbarino & Strahilevitz, 2004), and other newer CKM tools, such as weblogs or wikis (Wagner & Bolloju, 2005), could modify risk perceptions and results. Acknowledgements n CajaMurcia is grateFinancial support from Fundacio fully acknowledged. Appendix Internet knowledge (based on Novak et al., 2000)

   

I know tools for searching products on the Internet. I know how to nd in the Internet what I look for. Compared to other things I do with the computers, I consider myself as high skilled in using the Internet. Compared to other sports or hobbies, I consider myself as high skilled in using the Internet.

Internet preference (based on McKnight et al., 2002)

  

I like to explore new web sites. Among my colleagues, I am usually the rst to try out new web sites. When I have some free time, I often explore new web sites.

Risk preference (based on Chen & He, 2003)

   

I like to test myself every now and then by doing something a little risky. Sometimes I will take a risk just for the fun of it. I sometimes nd it exciting to do things for which I might get into trouble. Excitement and adventure are more important to me than security.

Online product perceived risk (based on different studies; Bauer, 1960; Mitchell, 1999, 2001) Which are your perceptions if you nd each of the following tools on a web site (Shared databases, Document repositories, Workow applications and Discussion forums)? Please provide a separate response to each of the tools.

The web site might not process correctly my purchase order.

ARTICLE IN PRESS
112 C. Lopez-Nicolas, F.J. Molina-Castillo / International Journal of Information Management 28 (2008) 102113

  

My personal data might be lost or use incorrectly. Time required to buy and obtain the product will be longer. Product delivery may last long or be incomplete.

Online purchase intention (based on Chen & He, 2003; Pavlou, 2003)

 

If this online retailer has the product I need to buy, I intend to buy it from the retailer. I would consider purchasing from this web site in the future.

References
Anderson, J. C., & Gerbing, D. W. (1988). Structural equation modelling in practice: A review and recommended two-step approach. Psychological Bulletin, 103(3), 411423. Bagozzi, R. P., & Yi, Y. (1988). On the evaluation of structural equation models. Journal of the Academy of Marketing Science, 16(1), 7494. Bauer, R. A. (1960). Consumer behavior as risk-taking. In Dynamic marketing for a changing world (pp. 389393). Chicago: American Marketing Association. Bierly, P., & Daly, P. (2002). Aligning human resource management practices and knowledge strategies: A theoretical framework. In C. W. Choo, & N. Bontis (Eds.), The strategic management of intellectual capital and organizational knowledge. Oxford: Oxford University Press. Borges, M. T., Almeida, J. P., Gomes, F., & Cabral, J. A. (2007). Knowledge management. An overview of European reality. Management Research News, 30(2), 100114. Bose, R. (2000). Knowledge management capabilities and infrastructure for e-commerce. Journal of Computer Information Systems, 42(5), 4049. Cases, A. S. (2002). Perceived risk and risk reduction strategies in Internet shopping. International Review of Retail, Distribution and Consumer Research, 12(49), 375394. Chen, R., & He, F. (2003). Examination of brand knowledge, perceived risk and consumers intention to adopt an online retailer. Total Quality Management and Business Excellence, 14(6), 677693. Chen, S. Y., & Macredie, R. D. (2005). The assessment of usability of electronic shopping: A heuristic evaluation. International Journal of Information Management, 25(6), 516532. Cheung, W. M., & Huang, W. (2002). An investigation of commercial usage of the World Wide Web: A picture from Singapore. International Journal of Information Management, 22(5), 377388. Chou, D. C., & Lin, B. (2002). Development of Web-based knowledge management systems. Human Systems Management, 21(2), 153158. Chung, W., Chen, H., & Nunamaker, J. F. (2005). A visual framework for knowledge discovery on the web: An empirical study of business intelligence exploration. Journal of Management Information Systems, 21(4), 5784. Conchar, M. P., Zinkhan, G. M., Peters, C., & Olavarrieta, S. (2004). An integrated framework for the conceptualization of consumers perceived-risk processing. Journal of the Academy of Marketing Science, 32(4), 418436. Das, S., Echambadi, R., McCardle, M., & Luckett, M. (2003). The effect of interpersonal trust, need for cognition, and social loneliness on shopping, information seeking and surng on the web. Marketing Letters, 14(3), 185202. Davenport, T. H., De Long, D. W., & Beers, M. C. (1998). Successful knowledge management projects. Sloan Management Review, 39(2), 4357.

Davenport, T. H., Harris, J. G., & Kohli, A. K. (2001). How do they know their customers so well? MIT Sloan Management Review, 42(2), 6373. DeSanctis, G., Fayard, A.-L., Roach, M., & Jiang, L. (2003). Learning in online forums. European Management Journal, 21(5), 565577. Doolin, B., Dillon, S., Thompson, F., & Corner, J. L. (2005). Perceived risk, the Internet shopping experience and online purchasing behavior: A New Zealand perspective. Journal of Global Information Management, 13(2), 6688. Du Plessis, M., & Boon, J. A. (2004). Knowledge management in eBusiness and customer relationship management: South African case study ndings. International Journal of Information Management, 24(1), 7386. Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18(1), 3950. Forsythe, S., & Shi, B. (2003). Consumer patronage and risk perceptions in Internet shopping. Journal of Business Research, 56(11), 867875. Garbarino, E., & Strahilevitz, M. (2004). Gender differences in the perceived risk of buying online and the effects of receiving a site recommendation. Journal of Business Research, 57(7), 768775. Gupta, A., Su, B., & Walter, Z. (2004). An empirical study of consumer switching from traditional to electronic channels: A purchase-decision process perspective. International Journal of Electronic Commerce, 8(3), 131161. Hansen, M., Nohria, N., & Tierney, T. (1999). Whats your strategy for managing knowledge? Harvard Business Review, 77(2), 106116. Hayes, N., & Walsham, G. (2001). Participation in groupware-mediated communities of practice: A socio-political analysis of knowledge working. Information and Organization, 11(4), 263288. Heinze, N., & Hu, Q. (2006). The evolution of corporate web presence: A longitudinal study of large American companies. International Journal of Information Management, 26(4), 313325. Kolbe, L. M., & Geib, M. (2005). Customer knowledge management. In Proceedings of the 38th Hawaii international conference on system sciences (HICSS). Kwan, M. M., & Balasubramanian, P. (2003). KnowledgeScope: Managing knowledge in context. Decision Support Systems, 35(4), 467486. Martin, W. J. (2001). The role of knowledge content in e-commerce. Journal of Information Science, 27(3), 180184. Maswera, T., Dawson, R., & Edwards, J. (2006). Assessing the levels of knowledge transfer within e-commerce websites of tourist organisations in Africa. Electronic Journal of Knowledge Management, 4(1), 5966. McKnight, D., Choudhury, V., & Kacmar, C. (2002). Developing and validating trust measures for e-commerce: An integrative typology. Information Systems Research, 13(3), 334359. Mitchell, V. W. (2001). Re-conceptualizing consumer store image processing using perceived risk. Journal of Business Research, 54(3), 167172. Miyazaki, A. D., & Fernandez, A. (2001). Consumer perceptions of privacy and security risks for online shopping. The Journal of Consumer Affairs, 35(1), 2744. Nonaka, I. (1994). A dynamic theory of organizational knowledge creation. Organization Science, 5(1), 1437. Novak, T. P., Hoffman, D. L., & Yung, Y. (2000). Measuring the customer experience in online environment: A structural modeling approach. Marketing Science, 19(1), 2242. Nysveen, H., & Pedersen, P. E. (2004). An exploratory study of customers perception of company web sites offering various interactive applications: Moderating effects of customers Internet experience. Decision Support Systems, 37(1), 137150. Ofek, E., & Sarvary, M. (2001). Leveraging the customer base: Creating competitive advantage through knowledge management. Management Science, 47(11), 4411456. Pavlou, P. A. (2003). Consumer acceptance of electronic commerce: Integrating trust and risk with the technology acceptance model. International Journal of Electronic Commerce, 7(3), 101134.

ARTICLE IN PRESS
C. Lopez-Nicolas, F.J. Molina-Castillo / International Journal of Information Management 28 (2008) 102113 Richard, M. O., & Chandra, R. (2005). A model of consumer web navigational behavior: Conceptual development and application. Journal of Business Research, 58(8), 10191029. Rollins, M., & Halinen, A. (2005). Customer knowledge management competence: Towards a theoretical framework. In Proceedings of the 38th Hawaii international conference on system sciences (HICSS) (pp. 110). Romano, N., & Fjermestad, J. (2003). Electronic commerce customer relationship management: A research agenda. Information Technology and Management, 4(2), 233258. Rowley, J. (2002). Eight questions for customer knowledge management in e-business. Journal of Knowledge Management, 6(5), 500511. Saeed, K. A., Grover, V., & Hwang, Y. (2005). The relationship of e-commerce competence to customer value and rm performance: An empirical investigation. Journal of Management Information Systems, 22(1), 223256. Salomann, H., Dous, M., Kolbe, L., & Brenner, W. (2005). Rejuvenating customer management: How to make knowledge for, from and about customers work. European Management Journal, 23(4), 392403. Shah, J. R., & Murtaza, M. B. (2005). Effective customer relationship management through web services. Journal of Computer Information Systems, 46(1), 98109. Shim, S., Eastlick, M. A., Lotz, S. L., & Warrington, P. (2001). An online prepurchase intentions model: The role of intention to search. Journal of Retailing, 77(10), 397416. Sitkin, S. B., & Pablo, A. L. (1992). Reconceptualizing the determinants of risk behavior. Academy of Management Review, 17(1), 938. Skyrme, D. J. (2001). Capitalizing on knowledge: From e-business to k-business. Oxford: Butterworth-Heinemann. Su, C. T., Chen, Y. H., & Sha, D. Y. (2006). Linking innovative product development with customer knowledge: A data-mining approach. Technovation, 26(7), 784795. Tsai, M. T., & Shih, C. M. (2004). The impact of marketing knowledge among managers on marketing capabilities and business performance. International Journal of Management, 21(4), 524530. Wagner, C., & Bolloju, N. (2005). Supporting knowledge management in organizations with conversational technologies: Discussion forums, weblogs and wikis. Journal of Database Management, 16(2), iviii. 113 Wang, S. (2001). Toward a general model for web-based information systems. International Journal of Information Management, 21(5), 385396. Westland, J. C. (2002). Transaction risk in electronic commerce. Decision Support Systems, 33(1), 87103. Workow Management Coalition. (1999). Terminology and glossary. Winchester: WfMC. Zack, M. H. (1999). Developing a knowledge strategy. California Management Review, 41(3), 125145.

Carolina Lopez-Nicolas (Ph.D., University of Murcia) is an assistant professor at the Department of Management and Finance at the University of Murcia (Spain). She holds a BA in Business Administration from University of Murcia and a BA Honours in Accounting and Finance in Europe from Manchester Metropolitan University. She has been a Visiting Professor at the Delft University of Technology in 2005 and 2007. Her current research relates to knowledge management, electronic business, electronic commerce and strategy. She has published on these topics in such journals as Journal of Knowledge Management, Journal of Enterprise Information Management, International Journal of E-Collaboration, and International Journal of Internet Marketing and Advertising.

Molina-Castillo (Ph.D., University of Murcia) is an Francisco Jose Assistant Professor of Marketing at the University of Murcia (Spain). He has a Masters Degree in Business and Foreign Trade, including a period of training at the Spanish Chamber of Commerce in Vienna, Austria. He received his BA in Business Administration from the University of Murcia and a BA Honours in Accounting and Finance in Europe from Manchester Metropolitan University. He has been a Visiting Professor at the Delft University of Technology in 2005 and 2007. His research interests focus on new product launch and electronic business. He has published on these topics in such journals as Telematics and Informatics and in the International Journal of Internet Marketing and Advertising.