Introduction

Knowledge is valuable. This comment can be found in over 6500 websites on the Internet using a Google search, and is one of the basic premises of Knowledge Management and the knowledge-based economy (e.g., Davenport & Prusak, 1998; Dragland, 2000; Lehrer, 2000; Harris, 2001). However, there is very little understanding of what knowledge is valuable, if all knowledge is valuable, or if only some knowledge is valuable (e.g., Dragland, 2000; Lehrer, 2000; Brockbank, 2004). Furthermore, there is little understanding of how the value of knowledge may differ depending on which perspective is taken. For example, the term 'knowledge is valuable' is often stated without further discussion of why it might be valuable or how the value is determined and by whom (e.g., the organization, the individual with the knowledge, or the individual seeking the knowledge). Without fully understanding how the value of knowledge is determined, it is somewhat simplistic to assume all 'knowledge is valuable.' It is also important to know how the value of knowledge may influence knowledge management processes, such as the willingness of a person to share the knowledge he/she may possess.

It is proposed within this paper that perceived value of knowledge (PVK) (the value that the individual places on his/her knowledge) may affect what knowledge people are willing to share. PVK has not been explicitly studied to date, but several authors have alluded to this construct and its potential impact on knowledge sharing (Leidner, 1999; Shariq, 1999; Augier et al., 2001). Value of products and services has been studied extensively within the marketing literature; however, it has not yet been studied or defined in any formal manner with respect to knowledge. In this paper, we present the development of the construct of PVK in terms of its definition, dimensions, and its measure. Accordingly, our first research question is:

1. What are the dimensions of PVK?

The value of knowledge may influence one's willingness to share said knowledge. The psychology literature suggests a negative relationship between value and sharing (Zinser et al., 1975; Bregman et al., 1984; Mosbacher et al., 1985; Kalman et al., 2002). However, this depends upon the sharer losing all or part of the value of the object as a result. Since the sharer of knowledge may not lose value from the act of sharing, then a different relationship may hold true for knowledge sharing (e.g., a positive relationship or a curvilinear relationship where an individual is more willing to share their knowledge as its value increases up to a point). Until we better understand how people determine the value of their knowledge and the impact on what knowledge people are willing to share, the relationship between value and sharing is difficult to predict and warrants further investigation. Therefore, our second research question is:

2. Does PVK influence a person's willingness to share his/her knowledge?

We conducted two studies aimed at addressing these research questions. First, a qualitative, exploratory study was conducted to examine, develop and operationalize the definition and dimensions of PVK. In this study, we also sought to understand an individual's intentions to share the differently valued knowledge with various targets (i.e., close colleagues, distant colleagues, and others within the organization via broadcast media like knowledge repositories, documents, mass e-mails). The second study was a quantitative study to refine and validate the measure of PVK and examine its relationship with sharing intention (Figure 1).

Figure 1.

Outline of the research design.

Full figure and legend (37K)

The contributions of this research are three-fold. First, we define the dimensions of the PVK construct and develop and validate the PVK scale. Second, we show that the PVK is positively associated with intentions to share knowledge. Third, we show that the PVK influences the underlying reasons for individuals' willingness to share their knowledge among three types of targets (i.e., the enablers and barriers to knowledge sharing). The following sections present a literature review on the value of knowledge, the methodology and results for the two studies, and implications of our findings.

Literature review

In this section, we first discuss what is meant by value and the different perspectives from which knowledge can be valued. Drawing on both marketing and knowledge management literatures, potential dimensions of knowledge value are identified.

Value has been well studied and defined in the marketing literature as '... an interactive relativistic preference or, more formally, as a relativistic (comparative, personal, situational) preference characterizing a subject's experience of interacting with some object' (Holbrook & Corfman, 1985; Socha, 1998). Zeithaml's (1988) study of how people distinguish value from quality and price found that the participants 'used the term [value] in many different ways, describing a wide variety of attributes and higher level abstractions that provided value to them. What constitutes value – even in a single product category – appears to be highly personal and idiosyncratic' (Zeithaml, 1988). Therefore, value is a perception (Day & Crask, 2000) and we would expect this to apply when the object being interacted with or considered is knowledge.

As value is perceptual, it is important to define the context of the perceptions. In terms of knowledge within an organizational context, perceived value can be from three different perspectives. First, the organization may value its employees' knowledge in terms of its influence on the employees' abilities to perform their jobs and for the organization to achieve its goals (i.e., achieve economic profits, competitive advantage).

Second, congruent with the marketing literature that examines consumer value (e.g., Holbrook & Corfman, 1985; Zeithaml, 1988; Socha, 1998; Day & Crask, 2000), it is possible for knowledge to be valued from the perspective of the knowledge seeker. It has been shown in the literature that value from the perspective of the seeker does influence knowledge-seeking behaviors, such that the more valued the knowledge, the more willing a person is to seek out the knowledge (Gupta & Govindarajan, 2000).

Third is the perspective of the knowledge owner/potential sharer. If the potential sharer is unwilling to share his or her knowledge due to its perceived value, it means little as to whether or not others perceive its value as high or low. This perspective has yet to receive empirical attention within the knowledge management literature and is the focus of the studies reported in this paper.

On what basis might one value knowledge? Several knowledge management researchers have offered suggestions. Gupta & Govindarajan (2000) argue that uniqueness is the basis of knowledge's value, such that unique knowledge is more valuable. Augier et al. (2001) state that the relevancy of the knowledge may determine its value; the more relevant the knowledge is to the problem at hand, the more valuable it is. Shariq (1999) adds tacitness and ease of replication to the list of possible determinants of value. These two relate to uniqueness and inimitability; the more tacit the knowledge is, the more difficult it is for others to copy it. The same holds for inimitability. Davenport & Prusak (1998) also argue that knowledge is valuable because it is close to action. This implies that the relevancy or usefulness of the knowledge for the individual's job might also be a factor in determining its value. These are possible dimensions for PVK; however, none of these have been tested empirically so it is uncertain if these dimensions are correct or complete.

A review of the marketing literature helped to identify additional possible dimensions for the value of knowledge from the perspective of the owner, not the consumer. Richins (1994) developed a possession rating scale, which had the following dimensions of value: utilitarian, enjoyment, interpersonal ties, identity, financial aspects, appearance-related, status, and spiritual. Some of these potential dimensions for PVK might not apply to knowledge since knowledge has different characteristics than material possessions. For example, knowledge is not a visible trait, thus appearance-related benefits of knowledge might not apply in a physical sense. However, one might argue that titles that denote the knowledge an individual possesses may substitute for the appearance-related benefits of knowledge (e.g., the title 'Dr.' for individuals who have a Ph.D. or medical-related degrees). Knowledge is also not typically directly 'bought' with money, so the financial aspect may also be somewhat different. Given the unique nature of knowledge, the first study, described below, empirically investigates how knowledge workers perceive the value of their knowledge.

Study 1

The first study was conducted to examine the meaning of PVK (PVK), and to understand how it might be associated with one's willingness to share one's knowledge. Since this was a new topic and had not been examined empirically to date, in-depth interviews were used to better understand PVK's dimensions and its impact on sharing knowledge.

Methodology

Telephone interviews were conducted with knowledge workers to address the research questions. (Details of the questions are excluded for space considerations, but they are available from the first author upon request.) The interviews took approximately 25–45 min, depending on the amount of discussion from the participant.

Sample

The sample for this study consisted of 20 employed knowledge workers working within North America. Knowledge workers were defined as: individuals who use information or knowledge as a primary input or output of their job. There were nine male participants and 11 female participants, with their age ranging from 24 to 55 years. The participants were a convenience sample from various industries (e.g., government, consulting, engineering, energy, information technology, and retail). Convergence on the findings was achieved so further interviews were not sought. The individuals were employees, not top management (examples of job titles included economic development officer, project manager for IS team, teacher assistant for special education children, quality inspector for software).

Analysis

The analysis of the interviews followed an iterative process, as suggested by Miles & Huberman (1994). First, the handwritten notes were checked against the recorded interview for accuracy. Then the interviews were listened to again, to listen for any emphasis given by the participant, and the emphases were marked within the handwritten notes. The handwritten notes were then examined to address the research questions.

To address the first research question (i.e., the dimensions of PVK), participants were asked to describe knowledge that they valued highly and low-valued knowledge. They were then asked to explain why the knowledge was high or low valued. Responses were analyzed to identify common themes. A theme was a potential dimension for PVK. The possible dimensions were then checked for redundancy and overlap to create a more coherent picture of PVK.

To address the second research question (i.e., the relationship between PVK and intentions to share knowledge) participants were asked about their willingness to share the differently valued knowledge with various target types (e.g., a close colleague or friend, a distant colleague, or broadcasting it to the whole organization) and why. The three different targets of sharing were included in this research as a way to control for some potential confounding effects since the reputation of the participant and nature of the relationship could influence PVK's relationship with intentions to share knowledge (e.g., Szulanski & Cappetta, 2000; Connelly & Kelloway, 2003; Thomas-Hunt et al., 2003). The willingness was quantified (as a yes, no, or some). Underlying reasons for sharing were then analyzed to identify trends and qualitative differences.

Results

Definition and dimensions of PVK

From the interview transcripts, seven overarching themes for PVK were identified. These were: usefulness, benefits from having the knowledge, the manner in which the knowledge was obtained (i.e., the source of the knowledge), uniqueness versus commonness, accessibility by others (which relates to uniqueness or commonness), tacitness, and information-like characteristic for low-valued knowledge. Each theme is described more fully below.

Usefulness of the knowledge was commented as a determinant of value by the most participants (85%). The more relevant and useful the knowledge was for the individual's work tasks, the more valued the knowledge was. For example, 'this knowledge keeps me out of trouble... . It makes me better and good at what I do.' A common comment was the following: 'It lets me meet the challenges of my job.'

Benefits from having the knowledge was the second most frequently mentioned dimension (50%). It included aspects such as being deemed the expert, gaining a sense of pride, deriving a competitive advantage, peace of mind, power, and this knowledge 'let's me keep my job and get paid.' The more benefits received from the knowledge, the more valued the knowledge was. Benefits were generally derived from simply having the knowledge (not necessarily using the knowledge). An exception to this was a comment made by a participant: Knowledge is power. The more you know and use it, the stronger an individual you become, but you have to use it to become stronger.

The third most common possible dimension was the manner in which the knowledge was obtained (45%), specifically, the origination of the knowledge. The more valued knowledge was obtained from experience, the school of hard knocks, a memorable teacher or mentor or even the person's family. Less valued knowledge was obtained from books, the Internet, an unreliable source, and education/school. For example, participants made the following comments: I obtained this knowledge through the school of hard knocks of personal experience. I can't put a price of it. You can't read it or find it on the Internet. Well, I have a lot of experience with this. I got this knowledge from my parents. My family is very, very important to me, so I value this knowledge.

The fourth possible dimension discussed by participants was uniqueness versus commonness (15%). Uniqueness was the extent to which the knowledge was distinctive from others' knowledge. Inimitability could be associated with this dimension. The more unique the knowledge was, the more valuable the knowledge was. The opposite of this possible dimension was commonness. In other words, the more common or well-known the knowledge was by others, the less valued the knowledge was.

Accessibility by others relates to uniqueness or commonness such that the more accessible the knowledge was to others, the less valuable the knowledge was. However, accessibility differs from uniqueness slightly such that it implies that others already have the knowledge (thus it is not unique). Only two participants (10%) commented on this factor. Accessibility was derived from comments such as the following: This knowledge is readily available to others. There's open access to it. Well, it's less valuable to me because it's more easily accessible to others.

Tacitness was commented by a couple of participants (10%) as a factor that described their high valued knowledge. In other words, their high-valued knowledge was 'hard to explain it for others to understand it' and that 'you just can't write down that stuff.' Interestingly, explicitness was not a determinant for the lower valued knowledge. Furthermore, the tacitness was not described as a defining factor for the value of the knowledge (e.g., the knowledge was not considered valuable because it was tacit); rather, it was described as a barrier to sharing the high-valued knowledge.

Finally, a few participants (15%) described their low-valued knowledge in terms of it being information-like. For example, one male participant commented that the low valued knowledge 'borders along information.' Another female participant commented that her low-valued knowledge constituted her 'book of useless information – the tidbit facts that are completely useless.' Based on Zack's (1999) argument that explicit knowledge is information, it would hold that this is the extreme end of the tacitness dimension (i.e., explicit knowledge), but this alignment of explicit knowledge with information is not a fully accepted premise. However, it should be noted that only a few participants commented on the tacitness or information-like aspects of the knowledge; whereas the other possible dimensions (e.g., usefulness, benefits, and source) were used more frequently.

Willingness to share

For the most part, people were willing to share their high- and low-valued knowledge with 71.7% of the responses being 'Yes, I'd be willing to share... .' However, there were different patterns to the willingness to share regarding the different target types (see Table 1 for a summary of the frequencies). With respect to high-valued knowledge, people seemed to be more willing to share this type of knowledge with a close colleague or friend than with a distant colleague. They were approximately just as willing to share their high-valued knowledge with the generalized other as they were with a distant colleague.

Table 1 - Frequencies of willingness to share responses for high- and low-valued knowledge and the targets (n=20).

Full table

With high-valued knowledge, people were willing to share 'some' of the knowledge with a distant colleague or the generalized other. This appeared to be a compromise between sharing and not sharing this high-valued knowledge. The bits of knowledge that they were willing to share tended to be the more basic, introductory knowledge so that they would not confuse the person with too many details. Alternatively, people were willing to share grand-level/overview knowledge (the big-picture) so that they would not waste their time giving an individual more knowledge than what that person needed to do his/her job.

With respect to the low-valued knowledge, people seemed to be just as likely to share this knowledge with a distant colleague as they were to share it with a close colleague or friend; however, they were less likely to share it with the generalized other. Interestingly, low-valued knowledge seemed to be an 'all or nothing' response as no one responded that they would share 'some of it.'

The effect of PVK was further understood when the underlying reasons for sharing were analyzed. PVK does seem to influence the underlying reasons for a person sharing or not sharing. As summarized in Tables 2 and 3, the underlying reasons for being willing to share (or not being willing to share) differed across the two values (high and low) and for each of the targets although there were some commonalities. It seems that for high-value knowledge, interpersonal factors are relevant for sharing (e.g., trust, interpersonal history, dislike) and that risk, confidentiality and ability to articulate the high-valued knowledge are barriers. Whereas, for low-value knowledge, entertainment, casual conversations or an explicit request for the knowledge are underlying reasons for sharing, and time and lack of need are barriers to sharing.

Table 2 - Summary of reasons for sharing and not sharing.

Full table

Table 3 - Sample profile.

Full table

Discussion of the findings

In order to move toward developing a short list of key dimensions to test in Study 2, the possible dimensions derived from the interviews were compared with the possible dimensions for PVK found within the literature. While usefulness/relevancy, tacitness, and uniqueness overlap with the dimensions within the literature, their emphasis differed between the literature and the interview results. The literature emphasized uniqueness and tacitness more whereas study participants emphasized usefulness/relevancy more. The participants from our study suggest that it is the usefulness of the knowledge that is the main driver of its value. One female participant even made the following comment: There is no such thing as low valued knowledge, because while it might not be useful in the present, it could be useful and valuable in the future. You just don't know [emphasis by participant]!

This quote illustrates how usefulness and value are linked. Only one participant of the 20 suggested that the uniqueness and its associated competitive advantage was the leading factor for his valuation of his knowledge. This knowledge is what gives me my greatest strengths, and gives me my competitive advantage. It makes me stand apart from the others.

New dimensions from the study were the source of the knowledge (which had not been discussed within the literature), and the benefits from owning the knowledge. These two dimensions were highlighted by the majority of the participants; therefore, they should be included as dimensions of PVK.

Tacitness and ease of replication, although suggested in the literature (Shariq, 1999) as possible dimensions, do not seem to be a major consideration for knowledge workers. It may be that these dimensions of value may be more important to the organization as a whole (e.g., unique knowledge in comparison to the market and the organization's competitors); however, for the individual they seem to be of less concern for determining the value. These two possible dimensions relate to the ability for others to acquire or imitate the same knowledge and are therefore related to the uniqueness of knowledge.

If one compares the proposed dimensions of value with the literature on the value of possessions and consumer value, then more similarities become apparent. For example, usefulness/relevancy relates to the utilitarian dimension of possession value (Richins, 1994), which is one of the top two dimensions of value. Uniqueness corresponds with identity (i.e., self-expression which is measured in terms of uniqueness from others: Richins, 1994). Benefits overlap with a couple of possession values (e.g., enjoyment (i.e., feelings of security), status (i.e., makes others think well of the owner, power): Richins, 1994). Finally, source could be construed in terms of effort and sacrifice in pursuit of developing the knowledge. In this case, source corresponds with the 'financial' dimension of possessions (although this dimension is solely defined in terms of money: Richins, 1994), or sacrifice (Zeithaml, 1988).

Based on the empirical results and literature discussed above, the PVK dimensions chosen for further investigation in Study 2 were usefulness, benefits, source, and uniqueness. Tacitness does not have any corresponding support from the value literature. Tacitness and information-like characteristics are not included as dimensions of PVK since, based on participants' comments, they do not seem to drive the perception of value; rather it appears to be a more after-the-fact aspect. Furthermore, tacitness and explicitness did not seem to reliably describe the value of knowledge (i.e., some explicit knowledge was considered valuable).

The chosen dimensions were developed into a scale (with items based on the interview transcripts), and were tested in the second study (see Appendix A for a list of the dimensions and items).

Study 2

A second study was conducted to develop and test the PVK scale, and its relationship with intentions to share knowledge.

Within the psychology literature, it has been found that the more valued an item is, the less likely an individual is to share that item. Specifically, the more money an individual had, the less willing s/he was to share it with another individual (Zinser et al., 1975; Bregman et al., 1984; Mosbacher et al., 1985; Cialdini, 2001; Kalman et al., 2002). However, if the individual has empathy towards the recipient, then s/he tends to share more (Zinser et al., 1975). Leidner (1999) contends that individuals will consider the value of the knowledge to themselves when they decide if they will share it. Thus, empirical investigation of the relationship between sharing and an individual's belief about the value of the knowledge is needed (Leidner, 1999).

While the psychology literature suggests that the relationship between PVK and intentions to share knowledge will be negative, the knowledge needs to be of some value before individuals are willing to invest their time and energy in sharing it as indicated by the first study. Therefore, two types of relationships will be tested: a linear relationship, and a curvilinear relationship (to test for an inverted U-shaped relationship).

Questionnaire development and initial testing

To develop the questionnaire-based measure and test its psychometric properties including reliability and face, discriminant, concurrent, convergent, and predictive validity, this study consisted of four phases: a card sort, an expert panel and a target population pretest, a pilot study and a main study.

The PVK measure met all of the requirements for the card sort (e.g., inter-rater reliability of a minimum score of 0.65 using Cohen's Kappa: Moore & Benbasat, 1991) and pretest (e.g., logical flow of questionnaire, typographical errors, is there evidence of question order effects: Czaja, 1997). These results supported the face validity, convergent and discriminant validity for the PVK measure (Churchill, 1979; Straub, 1989), and suggested minimal order effect in the questionnaire design (Czaja, 1997).

A pilot study with 46 knowledge workers was conducted to do a preliminary test of the reliability of the reflective indicators, the appropriateness of the formative indicators (for the Source PVK dimension), and the concurrent validity for PVK. The results suggested there was sufficient reliability (Cronbach's alpha >0.70); the indicators for the formative Source scale were appropriate; and all of the measures passed the criteria for concurrent validity. (More details of the phases of this study are available from the authors.)

Main study

The final phase of the second study involved testing the psychometric properties of the PVK measure with a larger sample to enable testing of unidimensionality, discriminant validity and predictive validity of the construct. The relationship between PVK and sharing intentions were also examined. The psychology literature suggests that the relationship between PVK and intentions to share knowledge will be negative; however, results of the first study indicated that the knowledge needs to be of some value before individuals are willing to invest their time and energy in sharing it. Therefore, two types of relationships were investigated: a linear relationship, and a curvilinear relationship where people share moderate knowledge readily, but share low- and high-value knowledge less (for different reasons, as suggested by participants in Study 1).

Methodology

Participants were given access to an online questionnaire, which was part of a larger study on knowledge sharing. The questionnaire contained the proposed PVK scale (see Appendix A for the items) and a measure for intentions to share knowledge.

Sample

The sample for this study consisted of 237 employed knowledge workers working within North America. Participants were recruited through cold calls to organizations and a North American online participant pool (with selection criteria of employment).

Analysis

To test the psychometric properties of the PVK Scale, several analyses were conducted (see Table 4 for a summary of analysis techniques used). First, each dimension was tested for unidimensionality (using PCA) and reliability (using internal consistency (IC; minimum of 0.80 sought), average variance extracted (AVE; minimum of 0.50 sought) (Barclay et al., 1995) and Cronbach's alpha (; minimum of 0.70 sought) (Churchill, 1979)) except for Source, which is a formative measure (measures of IC are inappropriate tests of formative measures: Diamantopoulos & Winklhofer, 2001; Diamantopoulos & Siguaw, 2002). Preference was given to the IC and AVE results as these analyses allow for indicators to have varying impact on the latent variable, whereas Cronbach's alpha assumes equal weights for each of the indicators (Barclay et al., 1995). Discriminant validity was tested by comparing the square root of AVE with the correlations of the other dimensions (Barclay et al., 1995).

Table 4 - Validity tests for construct measures.

Full table

To test convergent validity, the dimension-based measure of PVK was compared to a global measure of PVK. This was done through structural equation modeling using PLS due to the formative measure of Source and the concern of the predictive nature (not the fit of the model: Chin & Newsted, 1999).

Concurrent validity was tested by looking for group differences among high-valued knowledge, low-valued knowledge, and moderate-valued knowledge for each of the dimensions.

Finally, the relationship between PVK and intentions to share knowledge was tested using two analyses: a linear regression, and a curvilinear regression based on a quadratic equation as it is possible that the relationship between PVK is curvilinear (such that it is shaped as an inverted U). Averaged scores of the dimensions were averaged together to create a PVK score and were used for the regressions. Additional analyses were done to test PVK's relationship within intentions to share, given control variables (age, sex, and perceived management support).

Results: psychometric properties

To test the psychometric properties of the PVK Scale, the above analyses were conducted. All of the dimensions' measures had adequate IC (as measured by IC and AVE). Although two of the measures had somewhat low Cronbach's alpha values, IC and AVE indicators were given more weight since the items' loadings were not to be assumed to be equal (i.e., IC and AVE values are more appropriate indicators of reliability) (see Appendix A for the means, standard deviations, PCA and PLS loadings, and measures' IC, AVE, and Cronbach alpha scores).

For discriminant validity, the square root of AVE for each dimension was compared to the correlation table of the PVK dimensions and the global PVK measure. All measures passed the discriminant validity test such that r_{inter-construct} <AVE.

Convergent validity for the aggregated measure of PVK (based on the averaged scores for each dimension) was partially supported. PVK (aggregated) was significantly related to the global measure of PVK for each of the levels of knowledge value (high: r=0.577, P<0.001 low: r=0.618, P<0.001, moderate: r=0.591, P<0.001). This supports the convergent validity for the PVK (aggregated) measure. However, only three of the dimensions had acceptable PLS loadings, usefulness, benefits, and source. Uniqueness did not load well with PVK (high valued: =0.44, low valued: =0.38, moderate valued: =0.24). This implies that Uniqueness is a separate construct from PVK; the implications of this will be discussed in more detail in the discussion section. From this point hence, Uniqueness was excluded from the PVK construct and any analyses associated related to PVK.

Concurrent validity was tested for each of the dimensions using an analysis of variance (ANOVA) and least significant difference (LSD) post hoc test (=0.05). (Levels of value of knowledge were manipulated and each level of value was measured; the levels were: high, low, and moderate.) Concurrent validity was supported as all of the dimensions had significant F-scores (PVK: F_(2,669)=14.38, P<0.001; uniqueness: F_(2,669)=16.25, P<0.001; benefits: F_(2,669)=19.21, P<0.001; usefulness: F_(2,669)=25.19, P<0.001; and source: F_(2,669)=4.93, P<0.01). The LSD post hoc tests indicated that high-valued knowledge and moderately valued knowledge scored higher on all five dimensions than low-valued knowledge (P<0.05), and all but source had moderately valued knowledge significantly different from high- and low-valued knowledge.

The relationship between PVK and intentions to share knowledge was examined to test the predictive validity of PVK. Two linear regression analyses were done for each of the targets (i.e., close colleague, distant colleague, and generalized other). The first linear regression was a simple linear regression with PVK as the independent variable and intentions to share as the dependent variable (both variables were composite scores based on averages of the indicators; PVK's dimensions, excluding uniqueness, were averaged together to create a composite score). The second linear regression was a multiple linear regression with PVK as the first independent variable, and PVK² as the second independent variable (to test for a possible curvilinear relationship).

For all three targets, the linear relationship was significant and the quadratic relationship was not significant (close colleagues: F_(1,229)=20.093, P<0.001, r²=0.081; distant colleagues: F_(1,222)=10.658, P=0.001, r²=0.046; generalized others: F_(1,218)=13.266, <0.001, r²=0.057). This means the relationship between PVK and intentions to share was a simple relationship; there was no curvilinear relationship. For all three targets, PVK had a positive relationship with intentions to share knowledge (close colleagues: B=0.525; distant colleagues: B=0.457; generalized others: B=0.555). In other words, the more valuable the knowledge, the more likely the individual was to intend to share his/her knowledge. Since there was no curvilinear relationship, this implies that even the highly valued knowledge is associated with high intentions to share, contrary to the literature.

Overall, the new PVK scale demonstrated acceptable psychometric properties (see the final column in Table 4 for a summary of the psychometric property and validity tests).

Results: qualitative analysis of value and dimensions

Within the survey, participants were asked to consider knowledge that they valued highly, lowly, and moderately and to briefly describe the knowledge. The purpose of this was to prime the participants on high-, low-, and moderately valued knowledge before completing the PVK measure for high, moderate, and low from the concurrent validity test. The brief descriptions of the different types of knowledge allow one more type of analysis to be done to examine what kinds of knowledge are typically valued as high, moderate, and low. This analysis was done to explore whether or not another typology may explain the different levels of value (e.g., explicit versus tacit) or if PVK is a new and separate construct.

A total of 212 participants gave examples of their high-, low-, and moderate-valued knowledge. The following is a synopsis of the dimensions of the descriptions provided, along with a few examples. It should be noted that analysis/categorization of some of the examples provided was not possible because they were too brief or contextual to be understood fully.

For high-valued knowledge, several of the brief descriptions of the knowledge fell within the dimensions of PVK and uniqueness, while some did not. The descriptions that fall outside the identified dimensions of PVK may illustrate related constructs (e.g., uniqueness) or they may illustrate unidentified dimensions of PVK that future research should explore. The dimensions of the descriptions for high valued knowledge were the following:

• Use (47%): procedural 'How to...' (43), computer skills (23), customer/client needs/service (16), general use (5), legal (2), people skills (1), analytical skills (9);
• Benefits (0.5%): competitive advantage (1);
• Source (7%): Experience (6), Research (1), Friends (1), Education (professional) (5), Insight (1);
• Uniqueness (4%; 8);
• Other (17%): Confidential (8), Big Picture (11), Strategic (5), Structure and Politics (5), 'Important' (4), Demand (3).

Some examples of the descriptions of high-valued knowledge are: 'Intimated customer information,' 'Strategy,' 'Controlling behaviour problems/disapline [sic],' 'Knowledge directly impacting the business,' 'Knowledge that only I have,' 'Procedures to follow.'

For low-valued knowledge, several of the descriptions of the knowledge again fell within the dimensions of PVK and uniqueness, while some did not. The dimensions of the descriptions for low-valued knowledge were the following:

• Use (27%): computer skills (30), Routine (17), General (5), Not job related (5);
• Source (8%): Gossip (4), TV, Newspaper, Magazine (5), Internet (5), Email broadcasts (2);
• Uniqueness/Commonness (8%; 18);
• Other (9%): Daily Life, Personal, Hobbies (8), Small Picture/Low Level (4), Location of... (6), 'When to...'

Some examples of low-valued knowledge are: 'What's for lunch,' 'Where the bathroom are, etc.,' '2+2=4,' 'Knowledge that is not important to my job,' 'Magazine articles that I have read,' 'Idle gossip,' 'Using office equipment.'

Finally, for moderate-valued knowledge, classifying the descriptions of the knowledge was somewhat more difficult although they did follow the same patterns as the high- and low-valued knowledge, such that several of the descriptions of the knowledge again fell within the dimensions of PVK and uniqueness, while some did not. The dimensions of the descriptions for moderate-valued knowledge were the following:

• Use (36%): computer skills (23), Tricks of the trade (5), Procedural (36), Customer/client needs or information (13);
• Source (3%): Internet (2), Email (2), Experience (2);
• Uniqueness/Commonness (4%; 8);
• Other (4%): Confidentiality (2), Descriptive information (4), Corporate History (2).

Examples of the descriptions for moderate-valued knowledge are: 'Knowledge that could possibly be related to my job,' 'Knowledge that is not necessary but still valuable,' 'Company policy,' 'Tricks of the trade programming tactics developed by me and others,' 'Client lists,' 'Something that can be used to improve processes.' An interesting comment was made by one participant, 'There really is no moderate-value knowledge. It is all HIGH except as noted above [in low].' This comment is interesting as it reflects a comment made by a couple of participants in the exploratory study, who had commented that there was no low-valued knowledge for them (everything was of high value).

Results: qualitative analysis of value and alternative explanations

The descriptions were examined further to examine if there is a common 'type' of knowledge that is valued highly, lowly, or moderately. Additionally, the brief descriptions were examined to see if there may possibly be an alternative explanation for the relationship between PVK and intentions to share based on the type of knowledge that is high- versus low-value (e.g., tacit versus explicit, procedural versus declarative, or information versus knowledge: Nissen, 2002; Holsapple, 2003).

There does appear to be more examples of knowledge that are tacit in nature for high-valued knowledge (e.g., 'intricate knowledge of customer's history and individual needs,' and 'strategic planning'), but there are also several examples of knowledge that could be construed as more explicit (i.e., could be found in procedure manuals; 'entering information in database' and 'office procedures'). Examples of low-valued knowledge seem to be somewhat more explicit in nature (e.g., '2+2=4', 'how to send a fax'), but there are still examples that could be construed as more tacit in nature (e.g., 'knowledge that only my department knows in the organization,' 'how to conduct interviews').

The different levels of valued knowledge could also be described in terms of procedural knowledge (e.g., 'how-to'), and contextual knowledge (e.g., 'when-to' and 'why-to'); however, there is no clear trend among the three levels of value. The results suggest that there is no consensus on what 'type' of knowledge is highly valued and what 'type' of knowledge is less valued. It is evident that what an individual values highly does tend to be useful for the job, beneficial to the individual, and derived from experience or someone important to the individual. Furthermore, low-valued knowledge tends to be seen as less useful (perhaps for career advancing aspects or job performance critical aspects), less beneficial and from unreliable sources (such as office gossip). It appears as though PVK is an individual perception (as mentioned by Zeithaml, 1988), but that individuals may also have an overarching belief about the value of knowledge as a whole.

Discussion

PVK and its dimensions

The findings indicate that the PVK (for the individual who owns the knowledge) is operationalizable, and its dimensions are similar to the dimensions of value for possessions although they are not identical (possibly due to the unique nature of knowledge from tangible assets). The dimensions of PVK were found to be: usefulness, benefits, and source of knowledge.

Uniqueness was not found to be a reflective indicator of PVK, although it is likely related to the construct. These results contradict the literature on value of possessions and the knowledge management literature in terms of the role of uniqueness on value. Lepak and Snell (1999, 2002) offer suggestions that could help to explain this finding in that they suggest uniqueness is distinct from value in terms of human capital for firms. They contend that the value of human capital is based on Porter's value chain and is comprised of aspects that affect the effectiveness of production, whereas uniqueness is based on the resource-based theory of firms and is based on inimitability (Lepak & Snell, 1999, 2002). Therefore, it may be that uniqueness plays a different role as an antecedent to PVK or perhaps other possible antecedents to intentions to share, like psychological ownership (Pierce et al., 2001).

Alternatively, the roles of uniqueness and usefulness in determining value might be related to the internal knowledge market within the organization. If the internal knowledge market values and rewards unique knowledge, like a highly competitive environment where employees compete against each other or each individual is treated like a 'firm' (e.g., lawyers), then uniqueness might be a valid component of value. If the internal knowledge market values useful knowledge, then usefulness may be the more salient dimension to individuals. The results from this research suggest that the majority of individuals' work environments (i.e., internal knowledge markets) value useful knowledge more so than unique knowledge. Thus, usefulness would be more salient to individuals. In addition, unique knowledge may also have to be useful to have any value as markets are not likely to reward knowledge that is simply idiosyncratic with no relevant application. This line of thought, however, does not explain why benefits and source are salient determinants of value for the individual.

Finally, the qualitative descriptions of the high-, low-, and moderate-valued knowledge in Study 2 also supports Zeithaml's comment that value is an individual's perspective and can vary from individual to individual. This implies that it would be difficult for managers or researchers to pre-define 'valuable' knowledge as it is in the eye of the beholder. However, the dimensions do suggest that knowledge that is deemed useful, beneficial and from an important (or credible) source is more valuable.

PVK and intentions to share knowledge

PVK was found to be positively related to intentions to share knowledge although the psychology literature suggested a negative relationship. The findings suggest that the more valuable the knowledge is perceived to be, the more willing an individual is to share it. This could be considered as a threat to the predictive validity of this measure. However, the contradictory findings can be explained if we examine the underlying basis of the value of knowledge and the characteristics of knowledge.

The basis of value for knowledge (by the owner's perspective) is its usefulness, benefits, and the source of knowledge. Uniqueness appears to be a separate construct that may be related to the PVK, but, based on our findings, is not a dimension of valuable knowledge. Nor is it a major consideration in most individual's valuation of their knowledge as exemplified in the first study where two out of 20 individuals (10%) considered it a factor, and in the second study where eight out of 212 individuals (4%) described their valuable knowledge as unique and 18 out of 212 individuals (8%) described their low-valued knowledge as common. Thus, people are motivated to share useful and beneficial knowledge that was gained from experience. Sharing this knowledge does not diminish its value as it does not become any less useful or beneficial (or from a different source) due to the act of sharing. Rather, sharing this high-value knowledge may actually increase its value to the owner as it may give them more benefits (e.g., more respect from colleagues: Constant et al., 1994).

This logic does not follow with uniqueness since sharing unique knowledge may diminish its uniqueness. This study found that uniqueness appeared to be a separate construct from value. The findings may be because uniqueness potentially has a different relationship with sharing than does knowledge value. To further examine this possibility, post hoc regression analyses were run. The results of these regressions indicate that, indeed, uniqueness does have a negative relationship with intentions to share knowledge (both with and without PVK in the regression equation). Therefore, uniqueness of knowledge appears to be negatively related to knowledge sharing whereas PVK is positively related, supporting the logic that uniqueness affects sharing behavior differently than does PVK.

Implications for future research

Before identifying some possible avenues for future research, we would like to highlight some of the limitations of this research. First, the external validity of the results may be limited. For example, the samples for both studies were from English-speaking North America, and the associated national cultures or corporate cultures within North America may have influenced the findings here. In addition, given the variety of the individuals included in this research (e.g., industries, levels within organizations), it would be beneficial to examine this with a larger sample size to ensure generalizability. It is possible that PVK and uniqueness have different relationships with intentions to share knowledge within different contexts, and there may have been insufficient representation of those contexts within this sample (e.g., lawyers or highly competitive environments).

Second, this research only investigated knowledge value from the perspective of the individual possessing the knowledge. We did not examine the value of knowledge from a knowledge seeker's perspective. While PVK appears to be related to individuals' knowledge-sharing intentions within organizations, often times sharing occurs because someone is seeking the knowledge and has asked for it as illustrated in Study 1. Thus, while future research should include PVK as an important variable, this research has only examined half of the picture for knowledge sharing.

Future research should also examine the issue of how individuals determine the value of knowledge for someone else (i.e., a potential recipient). For example, Leidner (1999) proposes that individuals consider the value of the knowledge to themselves and the value to the recipient before deciding what to share with the recipient. Yet, to date, we do not understand how this valuation occurs.

Another possible perspective of value to study is from the organization's perspective. As suggested earlier, organizations may value knowledge that is unique more than knowledge that is useful due to the organizational need for economic profits for long-term survival and growth. If this is indeed the case, then future research should also examine how to leverage the valued knowledge of the employees into unique combinations of valued (useful) knowledge within the organization. Similarly, if the uniqueness of the individual's knowledge is negatively related to intentions to share, then future research should examine ways to overcome that barrier or examine ways to compensate for that barrier to enhance the organization's unique knowledge.

Another interesting finding that warrants further examination is the impact of the value of knowledge and the targets of sharing on the underlying enablers and barriers to sharing. Individuals are generally willing to share any type of knowledge (high- or low valued) with close colleagues and friends, but for different reasons (high-valued knowledge to help the friend or close colleague; low-valued knowledge for entertainment or idle conversation purposes). However, individuals are less willing to share their high-valued knowledge with distant colleagues due to distrust or dislike of the distant colleague or the distant colleague having less perceived need for the high-valued knowledge. On the other hand, individuals are willing to share their low-valued knowledge with distant colleagues for entertainment purposes (e.g., to participate in idle conversations with colleagues and share tidbits of knowledge that may be amusing).

Finally, the underlying reasons for sharing via broadcast media to the 'generalized other' (i.e., organization as a whole), were very different from the other targets. For high-valued knowledge, individuals were willing to share to meet a general need, to set a behavioral example, or to set standards. Barriers for high valued knowledge were time, ability to articulate and the lack of feedback of understanding (from the recipient(s) of the knowledge). Individuals were willing to share low-valued knowledge through broadcast media to help an individual who might need it, or for personal interest or fun. Most individuals were not willing to broadcast their low-valued knowledge as they saw the act as a waste of their time, and no one would have use for the knowledge.

Therefore, it appears that the target type is a very important factor, and should be included as a context factor in future research. Our research has shown that people differentiate between sharing with individuals directly versus sharing with unknown others through broadcast media (i.e., the generalized other). This finding is supported with individualistic social exchange theory (Ekeh, 1974) in that people are more willing to share directly with individuals as the benefits from sharing will be expected from the recipient.

Our findings also have an important implication for the developers of IT-based knowledge management tools and for practitioners who are implementing knowledge repositories and information systems to improve the sharing of knowledge. Less than 50% of the respondents were willing to share either their high-value knowledge or low-value knowledge with the generalized other (e.g., contribute to knowledge repositories, intranet, organizational libraries, etc.). On the positive side, not many people are willing to fill these repositories with low-valued knowledge; however, they are also not willing to share their high-valued knowledge. It is possible that the moderately valued knowledge (which was not examined in Study 1) is the fodder for knowledge repositories. If it is the moderately valued knowledge, and part of value is usefulness, are organizations enforcing mediocrity through knowledge repositories? Future research should investigate if this is indeed what happens, as well as investigate how to get people to contribute high-value knowledge.

Several ideas for future research that follow from our work have been discussed above. However, perhaps the key recommendation is that future studies of knowledge sharing should include the PVK construct in their analysis since it potentially affects when and why people share knowledge. Future research should expand upon this and investigate the other perspectives of knowledge as mentioned in the literature review (e.g., the perspective of the seeker, and the perspective of the organization).

Implications for practitioners

An obvious implication for practitioners is that they should focus on overcoming the barriers to sharing the high-valued knowledge within the organization (as listed as reasons for not sharing in Table 3). For example, organizations should examine the confidentiality restrictions of the high-valued knowledge and try to develop ways to mitigate the perceived risk in sharing this knowledge. This would lower the cost of sharing, leading to an expectation of positive benefits of sharing.

Another implication for practitioners focuses on the utility of knowledge repositories in assisting knowledge sharing among individuals. As mentioned earlier, if individuals are most likely to share moderately valued knowledge, then what are the implications for overall organizational performance and competitiveness? However, individuals do require communication media to be able to share their knowledge (either face to face or computer or technology mediated), so organizations should focus on enabling technologies instead of perhaps the broadcast media.

A third possible implication for practitioners revolves around the organizational structure. As noted in both the interviews and the survey, people are more willing to share valued knowledge with close colleagues (i.e., individuals with whom they work or socialize) than distant colleagues or unknown recipients of broadcast media. This suggests that if organizations want more knowledge sharing within the organization (across departments/silos/divisions), then allowing and giving opportunities for interpersonal relationships to form, communities of practice, or inter-departmental teams are critical. While these concepts are not new, this research suggests that they are even more important than previously believed due to the type of knowledge shared among close colleagues.

Finally, this research also supports the old adage of rewarding on the basis of sharing behaviors instead of expertise. If the organization rewards individuals for their expertise (especially if it is unique and useful expertise for the organization's success), Uniqueness becomes more relevant and salient, which is a barrier to intentions to share knowledge according to this research. Conversely, if the organization recognizes and rewards the sharing of useful knowledge, then this should minimize the negative impact of uniqueness and enhance the positive effect of PVK on knowledge sharing. Rewards and recognition can be done through performance appraisals, promotion, and informal and formal recognition (e.g., awards, bonuses).

Conclusion

This research was conducted to define the construct of PVK, to investigate whether or not it is associated with one's intentions to share his/her knowledge, and to examine the effect of the target on the sharing of knowledge. PVK was found to be based on three main dimensions: usefulness (i.e., relevance), benefits from owning the knowledge, and source of the knowledge. Uniqueness was found to be a separate, albeit related, construct, that was negatively related to intentions to share knowledge. PVK was found to be positively associated with willingness to share such that it appears to be directly related with intentions to share, but also may moderate the underlying reasons for sharing as illustrated in the qualitative results in Study 1 regarding enablers and barriers. Future research is required to test this relationship. Finally, this research confirms that the recipient influences what an individual is willing to share, such that individuals are more willing to share with a close colleague or friend than they are to a distant colleague or to the entire organization through broadcast media.

References

1. Augier M, Shariq SZ and Vendelo MT (2001) Understanding context: its emergence, transformation and role in tacit knowledge sharing. Journal of Knowledge Management 5(2), 125–136.
2. Barclay D, Higgins C and Thompson R (1995) The partial least squares (PLS) approach to causal modeling: personal computer adoption and use as an illustration. Technology Studies 2(2), 285–309.
3. Bregman NJ, Lipscomb TJ, McAllister HA and Mims M (1984) Sharing behavior: effect of denomination value and number. Journal of Genetic Psychology 144(1), 131–135.
4. Brockbank BJ (2004) War of Knowledge, from http://www.refresher.com/!warofknowledge.html, date accessed February 24, 2005.
5. Chin W (1998) The partial least squares approach to structural equation modeling. In Modern Methods for Business Research (MARCOULIDES GA, Ed), Lawrence Erlbaum Associates, Hillsdale, NJ, pp 259–336.
6. Chin W and Newsted PR (1999) Structural equation modeling analysis with small samples using partial least squares. In Statistical Strategies for Small Sample Research (HOYLE RH, Ed), Sage Publications, Thousand Oaks, pp 307–341.
7. Churchill Jr GA (1979) A paradigm for developing better measures of marketing constructs. Journal of Marketing Research 16(February), 64–73. | Article |
8. Cialdini RB (2001) Influence: Science and Practice. Allyn and Bacon, Boston, MA.
9. Connelly CE and Kelloway K (2003) Predictors of employees' perceptions of knowledge sharing cultures. Leadership & Organizational Development Journal 24(5/6), 294–301. | Article |
10. Constant D, Kiesler S and Sproull L (1994) What's mine is ours, or is it? A study of attitudes about information sharing. Information Systems Research 5(4), 400–421.
11. Czaja R (1997) Questionnaire pretesting comes of age. Marketing Bulletin 9, 52–66.
12. Davenport T and Prusak L (1998) Working Knowledge: How Organizations Manage What They Know. Harvard Business School Press, Boston, MA.
13. Day E and Crask MR (2000) Value assessment: the antecedent of customer satisfaction. Journal of Consumer Satisfaction, Dissatisfaction and Complaining Behavior 13, 52–60.
14. Diamantopoulos A and Siguaw JA (2002) Formative Versus Reflective Indicators in Measure Development: Does the Choice of Indicators Matter? Loughborough, U.K. (unpublished manuscript).
15. Diamantopoulos A and Winklhofer HM (2001) Index construction with formative indicators: an alternative to scale development. Journal of Marketing Research 38(May), 269–277. | Article |
16. Dragland A (2000) Knowledge as a capital asset from http://www.ntnu.no/gemini/2000-01e/38.html, date accessed February 25, 2005.
17. Ekeh PP (1974) Social Exchange Theory: The Two Traditions. Heinemann Educational Books Ltd, London.
18. Fornell C and Larker D (1981) Structural equation modeling and regression: guidelines for research practice. Journal of Marketing Research 18(1), 39–50. | Article |
19. Gerbing DW and Anderson JC (1988) An updated paradigm for scale development incorporating unidimensionality and its assessment. Journal of Marketing Research 25(2), 186–192. | Article |
20. Gupta AK and Govindarajan V (2000) Knowledge flows within multinational corporations. Strategic Management Journal 21, 473–496. | Article |
21. Harris RG (2001) The knowledge-based economy: Intellectual origins and new economic perspectives. International Journal of Management Reviews 3(1), 21–40. | Article |
22. Holbrook MB and Corfman KP (1985) Quality and value in the consumption experience: phaedrus rides again. In Perceived Quality (JACOBY J and OLSON J, Eds), Lexington Books, Lexington, MA, pp 31–57.
23. Holsapple CW (2003) Knowledge and its attributes. In Handbook on Knowledge Management, Vol. 1: Knowledge Matters (Holsapple CW, Ed), pp. 165–188, Springer Verlag: Berlin.
24. Jones TD (2004) Personal, Professional and Service Company Commitments in Service Relationships. Ph.D. Dissertation, Queen's University, Kingston, Ontario (unpublished).
25. Kalman ME, Monge P, Fulk J and Heino R (2002) Motivations to resolve communication dilemmas in database-mediated collaborations. Communication Research 29(2), 125–154. | Article |
26. Lehrer K (2000) Theory of Knowledge. Westview Press, Colorado, U.S.A.
27. Leidner DE (1999) Understanding information culture: integrating knowledge management systems into organizations. In Strategic Information Management (GALLIERS R, LEIDNER DE and BAKER B, Eds), pp 523–550, Butterworth Heinemann, Oxford.
28. Lepak DP and Snell SA (1999) The human resource architecture: toward a theory of human capital allocation and development. Academy of Management Review 24(1), 31–48. | Article |
29. Lepak DP and Snell SA (2002) Examining the human resource architecture: the relationships among human capital, employment, and human resource configurations. Journal of Management 28(4), 517–543. | Article |
30. Miles MB and Huberman AM (1994) Qualitative Data Analysis: An Expanded Source Book (2nd edn.). Sage Publications, Thousand Oaks, CA.
31. Moore GC and Benbasat I (1991) Development of an instrument to measure the perceptions of adopting an information technology innovation. Information Systems Research 2(3), 192–222.
32. Mosbacher BJ, Gruen GE and Rychlak JF (1985) Incentive value: the overlooked dimension in childhood sharing. Journal of Genetic Psychology 146(2), 197–204.
33. Nissen ME (2002) An extended model of knowledge-flow dynamics. Communications of the Association for Information Systems 8, 251–266.
34. Pierce JL, Kostova T and Dirks KT (2001) Toward a theory of psychological ownership in organizations. Academy of Management Review 26(2), 298–310.
35. Richins ML (1994) Valuing things: the public and private meanings of possessions. Journal of Consumer Research 21(3), 504–521. | Article |
36. Shariq SZ (1999) How does knowledge transform as it is transferred? Speculations on the possibility of a cognitive theory of knowledgescapes. Journal of Knowledge Management 3(4), 243–251. | Article |
37. Socha R (1998) Perceived risk as a determinant of perceived value. Retrieved July 14, 2003, from http://www.sbaer.uca.edu/Research/1998/MMA/98mma164.htm.
38. Straub D (1989) Validating instruments in MIS research. MIS Quarterly 13(2), 147–169. | Article |
39. Szulanski G and Cappetta R (2000) When credible sources share complex knowledge. The Wharton School, University of Pennsylvania (unpublished manuscript).
40. Tabachnick BG and Fidell LS (2001) Using Multivariate Statistics (4th edn.). Allyn and Bacon, Boston, MA.
41. Thomas-Hunt MC, Ogden TY and Neale MA (2003) Who's really sharing? Effects of social and expert status on knowledge exchange within groups. Management Science 49(4), 464–477. | Article |
42. Trochim WM (2000) The Research Methods Knowledge Base (2nd edn.). Atomic Dog Publishing, Cincinnati, OH.
43. Zeithaml VA (1988) Consumer perceptions of price, quality, and value: a means-end model and synthesis of evidence. Journal of Marketing 52, 2–22. | Article |
44. Zinser O, Perry JS and Edgar RM (1975) Affluence of the recipient, value of donations, and sharing behavior in preschool children. Journal of Psychology 89(2), 301–305.

Appendices

Appendix A

Refined PVK Scale Table A

Table A1.

Full table

Acknowledgements

We acknowledge the assistance received from Dr. Mark Nissen, Dr. John Edwards and the anonymous reviewers. We also acknowledge this research was partially funded by the Queen's Monieson Research Centre and the Social Sciences and Humanities Research Council of Canada. An earlier version of this research was presented at the 38th Hawaii International Conference for Systems Sciences at the Big Island, Hawaii.

About the author

Dianne P. Ford is an Assistant Professor in the School of Business and Economics at Nipissing University, North Bay, Canada. She received her Ph.D. (with a double major in Management Information Systems and Organizational Behavior) from Queen's University, Kingston, Canada. Her research interests include: knowledge management, cross-cultural issues, organizational culture, leadership, and trust. Her work has been presented at Canadian Psychological Association, Hawaii International Conference for Systems Sciences, and Administrative Sciences Association of Canada conferences, and she has published in IEEE Transactions in Engineering Management, Knowledge Management Research and Practice, and the Handbook on Knowledge Management.

D. Sandy Staples is an Associate Professor in the School of Business at Queen's University, Kingston, Canada. His research interests include knowledge-sharing issues and practices, virtual work (at both the team and individual level), and the effectiveness of information systems. Sandy has published articles in various journals and books including Organization Science, Information & Management, Information Systems Research, Journal of Management Information Systems, and KM Review.