F a c u l t y o f E c o n o m i c s a n d B u s i n e s s U n i v e r s i t y o f A m s t e r d a m
The effects of reward systems and social value
orientation on the motivation to share knowledge
08
Fall
Supervisor: Katinka Quintelier
Student: Annick Leeuwenberg - 5947324 Date: 22-07-2014
“An investment in knowledge always pays the best interest.”
Table of contents
1. Abstract 4
2. Introduction 5
3. Literature review 7
1. Knowledge sharing motivation 7
2. Reward systems 13
3. SVO 17
4. The conceptual model 19
4. Methodology 22
1. Sample 22
2. Measurements 25
1. Dependent variable: knowledge sharing motivation 25 2. Independent variable: reward systems 26
3. Independent variable: SVO 27
4. Control variables 29
5. Analysis and results 30
1. Data preparation 30
2. Reliability test 31
3. Descriptive statistics and correlations 33
4. ANOVA 35
6. Discussion 39
1. The effects of reward systems on knowledge sharing motivation 39
2. The effect of SVO 40
3. Additional results 42
4. Managerial implications 42
5. Limitations and future research recommendations 43
7. Conclusion 45 8. Reference list 46 9. Appendix 53 1. Matrix 53 2. E-mail 54 3. Survey 55
List of figures and tables
Figure 1: Conceptual model 21
Table 1: Representation of respondents 22
Table 2: Overview of items knowledge sharing motivation 26
Table 3: Overview of items SVO 28
Figure 2: SVO circle 28
Figure 3: Global steps of the analysis 30
Table 4: Mean, standard deviation and internal reliability of constructed
variables 32
Table 5: Correlation table 34
Table 6: Results of the ANOVA 36
1. Abstract
Reward systems are set up to (de)motivate people into a certain desired behaviour (Bartol and Srivastava, 2002; Lin, 2077a and b). Scholars found that reward systems indeed influence the motivation to share knowledge. Even so, the decision to share knowledge also depends on how individuals value the payoffs to others, which is also known as ‘social value orientation’ (SVO) (De Dreu and van Lange, 1995; De Lange, 1999; Galletta et al, 2003; Bridoux et al, 2011). Therefore, it is important to see what the interaction effects are of different types of rewards and social value orientation with respect to individuals’ willingness to share knowledge. This research uses a two level repeated measures ANOVA to examine the effects of individual and group rewards on the motivation to share knowledge among students. Results show that the type of reward affects the motivation to share knowledge. This is explained by the idea that the group formation itself, irrespective of the social value orientation of individuals, determines whether knowledge is shared. The results showed no significant results when testing the effect of SVO on the relation between reward systems and the motivation to share knowledge. Future research should continue to focus on the effect of different types of rewards, individual behaviours and knowledge sharing motivation.
2. Introduction
‘An investment in knowledge always pays the best interest’ according to Benjamin Franklin (1706 – 1790). He has an interesting thought as knowledge exists in organizations and individuals. It is shared across organizations from person to person (Felin, 2005; Foss, 2007). Knowledge is a scarce resource that is difficult to transfer (Grant, 1996; Tsai, 2002; Lindenberg and Foss, 2011). Perhaps you have played the Chinese whisper game when you were a child, where you needed to pass on words or a short story to your peers. At the end of the line the last person speaks up and it becomes apparent that the words became blurred (Biemann, 2006). This demonstrates the difficulty of transfering knowledge from one person to the other.
How are firms able to encourage their employees to share knowledge when it is difficult to transfer it? Linking the organizational level with the employee level is important to counter this problem as otherwise there would be no flow of knowledge within the firm. Scholars found that reward systems motivate employees to share their knowledge with colleagues within the firm (Felin, 2005; Foss, 2007; Lin, 2007a; Lin, 2007b). In this way employees create and appropriate value to the firm through sharing their knowledge, as it becomes knowledge of the organization (Coff, 1999; Grandori, 2001; Foss, Husted and Michailove, 2010; Felin et al, 2012).
Both organizational and educational literature found that reward systems motivate people individually or within a group to act in a certain behavior (Lin, 2007a and b). In addition, the level of reward (group or individually based) is an important factor in determining behavior (Lin, 2007a and b; Slavin 1980; Slavin 1983; Bridoux et al, 2011).
Scholars also found that the decision to share knowledge depends on how individuals value the payoffs to others, also known as their ‘social value orientation’
(SVO) (De Dreu and van Lange, 1995; De Lange, 1999; Galletta et al, 2003; Bridoux et al, 2011). There are two main types of SVO, individualists and reciprocators (Schwartz, 1994). The first type is where someone constantly makes the effort to maximize own benefit, where the latter is someone who constantly contributes to the group.
Research recommends to further research the effect of reward systems and the motivation to share with respect to individual factors and generalizability of findings. Therefore, this thesis will focus on the generalizability of results by examining students and their behaviour within a group and individually. This topic is also important for managers, apart from future research recommendations made by scholars. The need to know which reward system enables high flows of knowledge throughout the organization is growing with the explosion of advisory firms, such as accounting and consultancy firms (Riege, 2005; Baaij, 2014). Therefore, the research question of this thesis is,
What is the effect of reward systems on the motivation to share knowledge, taking into account that individuals differ in their SVO?
This research is structured as follows. The major concepts (knowledge sharing motivation, reward system and SVO) are discussed in chapter 3 from which the conceptual model is derived (shown in figure 1). Subsequently the methodology is explained. Chapter 5 provides the results of the analysis. The proceeding chapter explains the discussion of the results followed by the limitations of this research, the managerial implications and the future recommendations. This thesis ends with the conclusion.
3. Literature review
The introduction briefly quoted that knowledge is the best investment that pays the best interest. The example of the Chinese whisper game was given, from which it became clear that transferring knowledge is difficult. It became somewhat clear that it is difficult to encourage people to share their knowledge, however necessary as it enables firms to outperform their competitors. This section will provide more in-depth information on how knowledge is shared and encouraged within an organization. It explains the link between reward systems and SVO on the motivation to share knowledge. Please note that the context in this thesis is that of student groups, therefore each of the section will specifically address the educational literature on the topic.
1. Knowledge sharing motivation
Within each organization there are individuals, and where there are individuals there is knowledge(Gupta and Govindarajan, 2000; Felin, 2005). Knowledge is 'all that is known' within the organization; it is fuzzy and closely attached to the individual that holds it (Grant, 1996, p. 110; Ipe, 2003; p. 339). Moreover, it is a resource that helps the organization to create value with which it is able to achieve superior performance over others. Whenever firms outperform their competitors by creating (higher) value, they are able to obtain competitive advantage over others (Barney, 1986; Barney, 1991; Grant, 1996; Rumelt, 2003). Here, competitive advantage, with respect to knowledge, is ‘what the organization knows, how it uses what it knows and how fast it is able know something now’ (Prusak, 1997; Goh, 2002).
A crucial factor in achieving this advantage is the transferability of knowledge (Grant, 1996; Goh, 2002), which is argued to have enormous benefits for the
organization if managed properly. This is explained as organizational learning, where knowledge of individuals is shared with others within the organization in such a way that it becomes knowledge of the organization (Grant, 1996; Gupa and Govindarajan, 2000; Tsai, 2002). This sharing is important as otherwise the knowledge is of limited use for the organization as it resides in one individual (Ipe, 2003). When sharing with others, the knowledge becomes a public good, where all members of the organization are able to use and extend it through sharing their own knowledge (Grant, 1996; Ipe, 2003; Bridoux et al 2011).
Several scholars advise organizations to implement a system that coordinates an easier flow of knowledge throughout the organization by encouraging employees to share their knowledge and, thereby, contribute to organizational learning (Grant, 1996; Gupa and Govindarajan, 2000; Tsai, 2002; Cabrera and Cabrera, 2002). An example of such coordination is the use of a reward system, which serves as an incentive to motivate employees in engaging in knowledge sharing (Ipe, 2003). This will be elaborated further in the next section.
In order to create value and achieve competitive advantage for the firm it is important that its employees share knowledge. Knowledge sharing is defined as sharing ‘what is known’ from one individual to the other (who is able to understand and absorb the information) in a way that it is encouraged and benefited by the whole organization (Tsai, 2002; Ipe, 2003; Book, Zmud, Kim and Lee, 2005; Foss et al, 2010). Here, the 'what is known' concerns the information, know-how, processes and procedures of the organization. Without engaging in sharing knowledge, it becomes difficult for another person to acquire it. (Book et al, 2005).
Individuals also need to be motivated to share knowledge, which depends on the willingness of the individual to actually share their knowledge (Lin, 2007a;
2007b). It is seen as a voluntary act as someone needs to decide whether to share or not (Ipe, 2003). Lin (2007a and b) states that there are three factors that motivate someone to share knowledge: individual, organizational and technological factors. The next section will address the organizational factor. In the last section of this chapter, there will be a broader explanation about the motivation for an individual to share. This depends on (amongst others) individual preferences, which relates to the individual factor mentioned by Lin (2007a and b). The technological factors are beyond the scope of this research and are not further elaborated.
Knowledge sharing within the organization is subjective to coopetition, where it is about using collective knowledge to pursue common interest (Tsai, 2002, p. 180; Bridoux et al, 2011). In coopetition there is an element of both cooperation and competitiveness with respect to individuals. Both employees as well as the organization benefit from cooperation through the value creation that comes from sharing their knowledge (Cabrera an Cabrera, 2002). It increases the absorptive capacity (the way new knowledge is learned and implemented in the organization) and it increases innovativeness within the organization (Grant, 1996; Miller, Fern and Cardinal, 2007).
Competitiveness, on the other hand, means that individuals use the collective knowledge for personal gain in an attempt to outperform others (Williamson, 1975; Tsai, 2002, p.180). This results in individuals who, after gaining the collective knowledge, engage less and less in knowledge sharing. Here, knowledge sharing is a public good dilemma, meaning that individuals could take the collective knowledge for their personal benefit without contributing to it (Ipe, 2003; Bridoux et al, 2011). The factor that influences individuals to cooperate depends on the context (e.g. the reward system), known as external factors that motivate employees. There are also
internal factors that motivate them, for example power and reciprocity. Individuals tend to share knowledge to those who have more power and status (Ipe, 2003). In addition, the degree to which knowledge is shared also depends on the SVO of individuals (Grandori, 2001; Tsai, 2002; Felin, 2005; Foss, 2007; Wang, He and Mahoney, 2009; Bridoux et al, 2011). Both these aspects (reward systems and SVO) are elaborated further in the next sectionof this literature review.
Tsai (2002), Ipe (2003) and Lin (2007a and b) explain that knowledge sharing arises on the individual level (as stated in the previous paragraph), on a group level (in teams) and on the organizational level (to which individuals and groups contribute and where knowledge becomes an organizational asset). This thesis draws its focus on the encouragement for individuals to share on the organizational level. They explain that a group encourages individuals to share their knowledge with peers (Bartol and Srivastava, 2002). Individuals do so by face-to-face contact, where peers build a relationship of trust thereby encouraged to share their knowledge with others (also known as relational embeddedness). This is also found in the educational literature, where student groups are used to motivate students to share their knowledge (Schoenecker, Martell and Michlitsch, 1997).
In higher education, groups are used to encourage learning across students, motivating them to share their knowledge to peers (Schoenecker et al, 1997). The idea behind this is that in a team, whenever a student learns, it will be spread across the group (Slavin, 1983). Students develop their social, teamwork and leadership skills in the group. It is important that each member of the group has something to gain from the group (Cohen, 1994). Or that each of them perceives that the only way to achieve the goal is to work together (e.g. there should be positive goal interdependence). Similar to employees in an organization, students are encouraged to share knowledge
whenever there is a level of trust and whenever there is satisfaction within the group (Anderson, 2005). Trust and satisfaction are harmed whenever one group member dominates the group (Cohen, 1994; Schoenecker et al, 1997). Therefore, the purpose for the teacher is to provide a clear goal and structure for students in the group in order to remove the risks of lower trust and attractiveness within the group.
In that case, the dominator in that group loses peer status, which is how others in the group perceive that person and how it affects the overall trust and attractiveness of the group. This goal might be to encourage students to engage in either cooperative or collaborative learning. Both concepts are defined as students that work in small groups on an instructed and collective task where they share knowledge in order to achieve common learning goals (Slavin, 1980; Slavin, 1983, Cohen, 1994; Bruffee, 1995; Dornyei, 1997). They are rewarded and/ or receive recognition based on the extent to which they achieve their goals (Slavin, 1980; Cohen, 1994; Schoenecker et al, 1997). The difference is that cooperative learning is a way of learning where the teacher governs and evaluates the group. In addition, students rely on their teachers to do so. This type of learning is mainly used for primary and secondary school. Collaborative learning occurs when students are independent from the teacher such that it results in self-evaluation and self-governance. Moreover the competition shifts from the intragroup level (e.g. between individuals) to the intergroup level (e.g. between groups). Here, students operate on a higher educational level (e.g. higher education at on university level) by actively participating in verbal discussions with each other (Cohen, 1994). This higher level is what they learned in primary and/ or secondary school (Bruffee, 1995). Therefore, this thesis will focus on collaborative learning when analyzing the behavior of students.
There are other variables that play a role in the knowledge sharing process apart from the aforementioned reward systems and SVO. These variables are important but not the main focus of this thesis and are therefore included as control variables. Two of these variables are age and educational level. According to Bruffee (1995) these two variables affect the motivation to share knowledge sharing in the same way. Older students have more experience and knowledge. They tend to share knowledge more often when put together in a group, as they are motivated to learn from experiences of others. The same holds for the educational level, if a student has obtained a higher educational degree (for example a Master student who obtained its bachelor degree) than it is more likely that this student will share knowledge. More experience is, thus, believed to positively relate to knowledge sharing behavior as it assumes that students are motivated to learn from experiences from their peers. Even so, it is important to note that experience might also include that students have learned the benefit of group assignments, which motivates them to participate in sharing their knowledge with their peers in the group (Schoenecker et al, 1997). Another variable that influences the knowledge sharing process behavior is gender. Here, it is expected that females share their knowledge more often than males. The reason for this is that females are more helpful in nature (Connelly and Kelloway, 2003), at least in more intimate or less public contexts (Eagly and Crowley, 1986). In addition, the group size matters as well. The responsibilities in a group shift from the individual level to the group level whenever there are more members/individuals. Literature states that small groups (from 3-6 students) are best to motivate knowledge-sharing behavior (Cohen, 1994).
In short, it is important to research reward systems and SVO as it is a way of creating value and a way to understand how this value is created and appropriated
throughout the organization (Galletta et al, 2003; Lin, 2007). This section provided information on the motivation to share knowledge share. In this part of the literature it became clear that reward systems and SVO play an important role in knowledge sharing behavior. Therefore, the next section will provide an overview of reward systems in both the strategy literature as well as the educational literature.
2. Reward systems
Recent literature has elaborated widely on reward systems. Previous section stated that scholars advice organizations to implement a reward. The reason why it is important to research the effect of reward systems is that it is a way of coordinating knowledge sharing, thereby encouraging an easier transfer of that knowledge. This contributes than to a firm creating knowledge, which becomes valuable and a source of competitive advantage (Barney, 1991; Schlegelmilch and Chini, 2003).
Reward systems are a type of governance, which are used to handle relationships within the firm (Williamson, 2002). This means that they are used in order to encourage certain behavior of employees by the firm. Moreover this type of relationship is used to set a reward for the individual or a group, which is provided after a certain outcome is achieved (Kerr and Slocum, 1987). The reason to do so is that individuals need something in return whenever they share their knowledge (Bartol and Srivastava, 2002). Following Lin (2007), individuals will decide to share when the costs of so doing outweigh what they receive in return. In the educational literature, rewards are set in order to encourage students to work together in groups and share their knowledge of a course in order to increase learning (Slavin, 1980; Slavin, 1983; Dornyei, 1997). In other words, teachers set a reward in order to ‘handle’ their students by encouraging them to share knowledge. To conclude, both the
strategy and the educational literature have researched the link between the reward system and the motivation to share knowledge.
In the strategy literature Foss (2007) and Foss et al (2010) define this relationship as the knowledge governance approach. Here they consider that each governance mechanism (including reward systems) has an effect on knowledge processes, such as sharing, retaining and cresting knowledge (Foss, 2007, p. 13). When researching this approach, it is important to note that sharing (especially the knowledge; Gant, 1996) is surrounded by causal ambiguity, bounded rationality and uncertainty (Williamson, 2002; Foss, 2007; Foss et al, 2010). Causal ambiguity means that it is unclear how someone achieved a resulting outcome (Szulanski, Capetta and Jensen, 2004). Moreover, it is about the things that form an unknown power block against sharing knowledge, thereby lowering the transferability of that knowledge (Simonin, 1999). Bounded rationality is that individuals have a boundary to what they are capableto understand. It is a cognitive limitation of an individual who is not able to make an informed decision about what to do (i.e., in this case the motivation to share knowledge) (Augier, Shariq and Vendelo, 2001). In addition, uncertainty is where individuals do not know which knowledge exists and whether the co-worker or co-student is willing and able to share the knowledge (Foss et al, 2010).
Scholars are, apart from these restrictions, also unsure about the nature of the relationship between the reward system and the knowledge sharing process. There is established that there is a relation between both, however they found both positive and negative results (Bartol and Srivastava, 2002; Foss, 2007; Lin, 2007; Foss et al, 2010). The reasoning behind the positive side is that individuals feel reinforced by receiving a reward that is linked to a firm’s values and beliefs, which in turn contributes to an increase in knowledge sharing (Kerr and Slocum, 1987; Bartol and
Srivastava, 2002; Lin, 2007). The negative result assumes that self-interested individuals, motivated through rewards, cause a demotivating effect to reciprocal behavior of others. In turn, the reciprocators within the firm are reluctant to share their knowledge in the future (Bartol and Srivastava, 2002).
Within this relationship, the educational literature explains two different types of reward systems. There is the individual reward and the group reward system (Slavin 1980; Webb, 1982; Slavin, 1983; Cohen, 1994; Bartol and Srivastava, 2002). The individual reward is where students who work independently or in a group receive a reward based on their individual performance. Even though individual rewards usually motivate students to work alone (Cohen, 1994), they might be motivated enough to work within a study group with familiar co-students, with whom they have formed a group in the past (Webb. 1982; Johnson et al 1988; Cohen, 1994). This effect is even stronger when the teacher advises the students to help each other to
learn (Dornyei, 1997). This might be explained by the fact that there is a positive relation between giving help and asking for help and the overall performance for an assignment (Webb, 1982).
For the group reward, students collaborate with each other or contribute to the group individually. For their performance they receive a reward for the entire group and so it is the same for each member. Whenever a group has the task to complete an assignment as a group, each member is given the incentive to increase the group performance (Slavin, 1980; Slavin, 1983; Cohen, 1994). Here each member is likely to share his or her knowledge to contribute to collaborative learning, resulting in a higher performance (Cohen, 1994).
Within these two types of rewards there are four examples of how rewards are used to encourage students towards sharing knowledge. Each type of incentive
considers either group or individual assignments for which either a group or individual reward is set. In this thesis we make a distinction between the individualistic incentive, the study group incentive, the cooperative and the competitive incentive (Slavin 1980; Webb, 1982; Slavin, 1983, Johnson et al, 1988). Each of these incentives is further explained and is summarized in the appendix 1.
Individual rewards exist in both the individualistic and study group incentive. With the individualistic incentive, students study individually and are rewarded based on their own performance (Slavin, 1983; p. 429). The other individual incentive is the study group incentive, which is based on the article of Webb (1982) and Cohen (1994). Here there is a study group of individuals that work together but are rewarded individually. The idea behind this is that students who worked together with each other in the past in the same group gather themselves to study for an individual assignment (Cohen, 1994). The effect of the study group is strengthened whenever the teacher advised students to form one group (Slavin, 1983). In addition, there is considered to be an overall incentive to reciprocate as providing help and receiving help by other students increases performance, regardless of the assignment (Webb, 1982).
Group rewards contain either a competitive and cooperative incentive. The competitive incentive is where the individual competes with others in the group to eventually obtain a group reward based on the best performers (Slavin, 1983, p. 429). Everyone in the group receives the same assignment, so individuals compete with respect to who performs best on that assignment. Here there is assumed to be less knowledge sharing and collaborative learning, as students tend to withhold information due to the competition (Slavin, 1980; Webb, 1982, Slavin, 1983; Cohen, 1994; Williamson, 2002). Nevertheless, it is likely that there is still a knowledge pool
from which each member could draw knowledge for the benefit of that person only (and not for the group). The other example is the cooperative incentive. Here students work together to achieve a group reward for a group assignment (Slavin, 1983, p. 429). Students are best enforced to work together as there is reward interdependence (Cohen, 1994), meaning that how well each member of the group performs determines the reward for that entire group. The link between helping others and obtaining help in the group immediately relates to higher performance, regardless of the type of person (Galletta et al, 2003).
Therefore in general a group of students that receives a group reward is more motivated to work together in order to share knowledge and increase collaborative learning; therefore the following hypothesis is formulated;
H1: Students that receive a group reward are more motivated to share knowledge than students that receive an individual reward.
Apart from reward systems, there is also the issue of differences between individuals (e.g. their SVO). Therefore the next section will elaborate on the relationship between these differences and the motivation to share knowledge.
3. SVO
SVO is a way to explain the differences in behavior between individuals. It is defined as the way individuals behave differently towards outcomes that are relevant for themselves and others in a group (De Dreu and van Lange, 1995; De Lange, 1999; Galletta et al, 2003; Bridoux et al, 2011). It is important to include this in research, as this is a way to examine differences across individuals.
Recent literature has indicated several similar types. Following Galletta et al (2003) and Bridoux et al (2011), the most common ones are reciprocator and
individualistic characteristics. Schwartz (1994) indicates them as self-transcendence and self-enhancement. Murphy, Ackerman and Handgraaf (2011) define them as individualistic and pro-social. The individualistic type (or self-interested individual) is someone that only cooperates for maximizing own personal benefit. This person will only benefit within a group, whenever the performance of that person increases. The pro-social type (or reciprocator) is someone who aims to increase the outcomes for the whole group, sometimes even when that means a lower benefit for that person. This person will do so by constantly helping others. Pro-socials share a feeling of collectivism, where it is important that other members in the group also aim to maximize the performance of that group (De Lange 1999; Galletta et al, 2003; p. 3-4; Bridoux et al, 2011, p. 713 – 714).
Similar types are found in the educational literature as well. Here they are linked to the cooperative, the competitive and the individualistic incentive on which the grading of an assignment is based (Slavin, 1980; Webb, 1982; Slavin, 1983; Johnson et al, 1988; Cohen, 1994; Dornyei, 1997). These incentives are implemented for different study groups in order to derive a certain behavior of students within certain configurations of different groups. This is explained in the previous section and will be combined with reward systems in the next section.
There is also a relationship found between helping and giving when performing in a study group for an individual assignment (Webb, 1982; Cohen, 1994). Students increase their performance both by either getting help for themselves or by helping others. This is interesting as this provides an incentive even for individualistic students to cooperate with others, as they would simply increase performance.
Due to the fact that it is in the nature of reciprocators to help others, it is believed that they probably will engage more in knowledge sharing behavior than individualists. Therefore, the following hypothesis is;
H2: Reciprocators are more motivated to share knowledge than individualists.
Both the effects of reward systems and SVO’s on knowledge sharing behavior are now explained. The following section will elaborate on the conceptual model that is derived from the literature.
4. The conceptual model
In the previous sections the different reward systems and SVO were discussed. This part will combine the individual and group reward (including the four incentives) with the individualist and reciprocator types. These combinations form, in the end, the conceptual model of this thesis, which will be tested later on.
In the section on reward systems the hypothesis is that students are more motivated to share knowledge if their reward is based on the group, so on the performance of themselves and their peers combined. In the section on SVO, we argued that reciprocators are more motivated to share knowledge than individualists. Combining these two would implicate that whenever students receive a group reward, they are more motivated to share knowledge, especially whenever they are reciprocators. However, this thesis would argue otherwise.
It is in the nature of reciprocators to always share knowledge regardless of the type of assignment (Cohen, 1994, Galletta, 2003; Bridoux et al, 2011). This is in contrast to individualists, who need encouragement to share knowledge as they normally focus on their personal benefit (Cohen, 1994). They will engage in
knowledge sharing activities only whenever there is a group-based reward, because this aligns goals of each individual in the group (Johnson et al, 1988).
In addition, reciprocators are discouraged in group-based assignments to share knowledge, as there are problems related to free riding. Free riding is where members of a group benefit from efforts made by the group, without contributing to the combined effort themselves. The effect for reciprocators is that they, in turn, participate less and less in knowledge sharing activities (Dornyei, 1997). This effect could be reduced depending on the social acceptance coming from the efforts to participate in the common good. In turn, this depends on the degree to which there are reciprocators in the group (Bridoux et al, 2011).
Therefore, the following hypothesis becomes:
H3: Students are more motivated to share knowledge whenever their reward is group-based, and this effect will be larger when they are individualists.
The conceptual model of this thesis is shown in figure 1 below. Now that there is a conceptual model, it is important to collect data in order to analyze whether the hypotheses are found. The way this thesis derives its data is, therefore, explained in the next chapter.
Figure 1 - Conceptual model Knowledge sharing motivation Control variables -‐ Age -‐ Gender -‐ Education -‐ Group size SVO -‐ Individualist -‐ Prosocial H1 H3 H2 Reward systems Individual reward Group reward
4. Methodology 1. Sample
The students of the Faculty Economics and Business from the University of Amsterdam were taken as a sample. The total group of students consisted of Bachelor, pre-master, Master, executive and ‘course’ students. In total this contributed to a population of 8,9971 people, with 4544 regular students and 4453 executive students. The expected response rate was 200. The population was approached through email (of which a database exists that is available to the writer of this thesis), personal contact and social media. The survey was made in Qualtrics and was sent on the 22nd of April 2014. The reminder for filling in the survey was sent two weeks later on the first of May 2014. The e-mail that was sent to the students is shown in appendix 2
Each student was asked to fill in a survey that included scales for the main variables (reward systems, SVO and knowledge sharing behavior. In addition, the demographics were asked (e.g. gender, age, education, specialization and GPA). The survey that was sent to the sample is shown in the appendix 3.
After distributing the survey the response rate was low (in total 514 entries, which relates to 5.7% of the total population). Considering only general students contributed to a response rate of 12.4%. From this data, the missing values of respondents that stopped or did not fill in all the knowledge sharing questions were deleted. In addition, for the final analysis the missing variables of SVO were removed as these increase errors in the model. The values for altruistic and competitive characteristics in SVO were omitted from the sample, as the goal of this thesis is to research the effects of only individualists and prosocial students. This resulted in a sample of 165 entries.
1 For the academic year 2013 – 2014, 4155 students were enrolled at the FEB. Data is obtained from the UvA database.
In this group the distribution of males and females was almost equal (53% male and 47% female). The average age of the sample was 24 years with a standard deviation of 4.139 years. The majority of respondents followed a Master program (63.1%), the residual were Bachelor students including pre-master students (36.9%). On average the size of a study group consists of 3.71 people with a standard deviation of 1.042. In other words, most groups were between 2 to 5 people. Table 1 shows an overview of the representation of respondents.
Table 1 – Representation of respondents
Demographic characteristics Frequency Percentages Gender Male 78 52.3 Female 71 47.7 Age 18 4 2.7 19 9 6.0 20 7 4.7 21 15 10.1 22 19 12.8 23 25 16.8 24 19 12.8 25 15 10.1 26 8 5.4 >27 27 18.6 Education level Bachelor 55 36.9 Master 94 63.1 Group size 1 1 0.7 2 22 14.8 3 34 22.8 4 56 37.6 5 34 22.8 6 2 1.3
Social value orientation
Prosocial 77 51.7
2. Measurements
The scales of each construct (knowledge sharing process, reward systems SVO) are tested in earlier literature. These scaled have been validated and proven to score high on reliability and validity. This is important, as this increases the reliability of validity of this research as well. Here, they are taken together in a different context (e.g. of the educational environment) in order to analyze the hypotheses. In addition, the scales of reward systems needed adjustments in order to research students that receive a group and/ or an individual reward. Therefore, in this section each scale that is used and adjusted for the research is discussed.
1. Dependent variable: knowledge sharing motivation The motivation to share knowledge is measured using the scale from Lin (2007a). This scale consists of 14 statements. Each statement is adjusted to the situation of students. Whenever the statement was not relevant for the context of this research it was removed. An example of this is the sub variable ‘expected organizational reward’. The final list of statements was asked twice, one for the knowledge sharing motivation for the group reward and the other for the individual reward. So, each student indicated to what extend they (dis) agreed with statements about their motivation to share knowledge. Respondents were asked to judge each statement on a 6 point Likert-scale. So they needed to state the extent to which they agreed (from disagree =1 to agree=6). Table 2 shows an overview of the item list for the motivation to share knowledge.
Table 2 – Overview of items knowledge sharing motivation
Reciprocal benefits 1. I feel that I strengthen my relationship with fellow students by helping them with my knowledge.
2. A higher grade was received because I shared my knowledge of the course.
3. Fellow students in the group helped me with their knowledge for the benefit of
the group.
Enjoyment of helping others 1. I enjoyed sharing my knowledge with fellow students.
2. I had the expertise to provide my fellow students with valuable knowledge for the course.
3. I enjoyed working with my fellow students in a group.
Knowledge self-efficacy 1. I was confident in my ability to provide knowledge that other students experienced as valuable.
2. It did not make a difference whether I shared my knowledge with fellows (recoded). 3. Other students have more valuable knowledge (recoded).
Attitudes towards knowledge sharing 1. It felt good to help someone with my knowledge.
2. Sharing knowledge with fellow students is pleasurable.
Knowledge sharing intentions 1. I intend to share my knowledge more frequently with fellows in the
future.
2. I will try to share my knowledge with other fellow students.
3. I will always share my knowledge with others.
2. Independent variable; rewards systems
Reward systems consist of the following two levels: group and individual rewards. Both of the two reward systems are one of the independent variables. Students needed to choose one course for which they both received a group (e.g. a paper and/ or presentation) and an individual reward (a paper or an exam). They were asked which course they followed in the last block and to what extend the final grade was based on the individual and the group reward. After that they were given the list of items on knowledge sharing motivations (as elaborated in the last section). They filled in that list for both rewards separately in order to be able to analyze the differences between the two.
3. Independent variable; SVO
The second independent variable is SVO of students. The hypotheses derived in the previous chapter indicated two different relationships. The first hypothesis was about the relationship between SVO and knowledge sharing behavior where prosocial students are expected to share more knowledge. The second hypothesis in the conceptual model assumed that SVO is a moderating variable on the relation between reward systems and the motivation to share knowledge.
With respect to this variable the respondents were introduced to the Murphy slider from Murphy, Ackermann and Handgraaf (2011). After the introduction in the survey, they were asked to make a decision on how to distribute an amount of money amongst them and a random other person. So, for example in the first question the respondent needs to decide to either provide the same amount of money to themselves and the other person, or to decide whether the other person should receive less. The idea is that whenever a student prefers the best outcome for themselves and the other (random) person, he/she is prosocial. If he/ she is an individualist, than the purpose is to benefit the best for themselves, which results in the highest amount of money for themselves and the lowest amount of money for the other person. For each question, the student had a total of nine options to choose what amount of money they would receive (marked as ‘you receive’) and what amount of money the random other person receives (marked as other receive)’. Table 3 shows the first six items of the Murphy slider including the options that the student received per question.
On the basis of participants’ answers an analysis is made using the means of each score. The means are put into a formula that calculates the corners of a circle. This circle is used in order to label each respond to a SVO type. The label of each
student, either ‘prosocial’ or ‘individualist’, corresponds to the size of the corner on that circle. Figure 2 shows how this corresponds to a circle.
Table 3 – Overview of items SVO SVO 1 2 3 4 5 6 7 8 9 1 You receive 85 85 85 85 85 85 85 85 85 Other receive 85 76 68 59 50 41 33 24 15 2 You receive 85 87 89 91 93 94 96 98 100 Other receive 18 19 24 28 33 37 41 46 50 3 You receive 50 54 59 63 68 72 76 81 85 Other receive 100 98 96 94 93 91 89 87 85 4 You receive 50 54 59 63 68 72 76 81 85 Other receive 100 89 79 68 58 47 36 2 15 5 You receive 100 94 88 81 75 69 63 56 50 Other receive 50 56 63 69 75 81 88 94 100 6 You receive 100 98 96 94 93 91 89 87 85 Other receive 50 54 59 63 68 72 76 81 85
On the website of the Murphy slider measure, there is a syntax made available in order to compute the values for SVO in SPSS. Please note that in the circle there are four different SVO’s, however this thesis draws its focus only on the two main characteristics (e.g. prosocial and individualist). The next section explains how the control variables are tested.
4. Control variables
The control variables in this model are age, gender, education and group size. These are indicated in the literature review as variables that influence knowledge sharing behavior, but are beyond the scope of this thesis. In the questionnaire these variables are asked at the last part of the survey under ‘demographics’. The control variables are included in the analysis, in order to control for them in the conceptual model.
The next chapter will provide the analysis and results for this thesis. It explains how the data obtained from the survey is processed in order to test the hypotheses.
5. Analysis and results
The analysis for this thesis is done on a step-to-step basis. The following figure, figure 3, indicates which global steps are taken to conduct the analysis. This part of the chapter will explain each part separately in order to fully elaborate on how the analysis is done.
Figure 3 – Global steps of the analysis
1. Data preparation
The data was retrieved from the Qualtrics website. As stated in the previous chapter, it contained 514 entries. For this part of the analysis it was central to prep the data into a form that it is ready for further analysis. This started with treating missing values and proceeded with computing variables needed for the analysis. This is about constructing the group reward, individual reward and the SVO variables. These two stages are shortly discussed here.
Data prep
Factor
analysis and
reliability
test
Descriptive
statistics and
correlation
table
ANOVA
From these entries (e.g. the 514 respondents), there were a lot of missing values. In the methodology chapter, the researcher explained that individuals with missing values for knowledge sharing motivation questions were removed, the values of ‘other’ for the education variable were also removed as these consisted of a group of outcomes, which were not traceable to one of the main educational levels. This left a total of 151 outcomes.
The second part of data preparation consisted of computing the variables needed for the analysis. At first, I separated all the statements for the group rewards into the different sub-variables stated by Lin (2007), which were made into general means. The same was done for the individual reward. The variable of SVO was made using the syntax provided on the website of the Murphy slider. The control variable age is grouped into general students (all ages under 28 were marked as a ‘0’) and executive students (all students aged 28 or higher were marked as ‘1’). All other control variables remained the same. All uncategorized answers were removed, which relates to the answer of ‘other’ for the educational level, in order to reduce the number of errors.
The second step after the data preparation is to test the data in terms of a reliability test (through Cronbach’s Alpha).
2. Reliability test
A reliability test is done in order to conclude whether the sub variables explain the main variable. In other words, whether the (sub) variables add up to what they are supposed to measure. Following Nunally and Bernstein (1994, p. 52), the internal consistency reliability measured through the Cronbach’s Alpha is needed in order to assess whether each sub variable is measuring the content.
The reliability levels of both types of rewards are tested via the use of Cronbach’s Alpha. For the group reward the Cronbach’s Alpha is 0.104 without removing any sub variables. The results indicate that removing the sub variable ‘knowledge self-efficacy’ leads an increase of the Cronbach’s Alpha to 0.684. Therefore this sub variable will be omitted for further analysis. The individual reward shows an increase of the Cronbach’s Alpha from 0.300 to 0.802, when knowledge self-efficacy is removed. The overall model shows a Cronbach’s Alpha of 0.736. Following Hirkin (1998), whenever there are still multiple sub variables left it is safe to conclude that the model and variable is high on reliability. Therefore, this sub variable (knowledge sharing self-efficacy) was excluded from the model. The following table shows an overview of all the values.
Table 4 – Mean, standard deviation and internal reliability of constructed variables (N=149)
Scale* Cronbach’s Alpha Mean (SD) Items
Types of reward 0.736 22
Individual reward 0.786** 17.56(6.544) 11
Group reward 0.669** 20.72(3.120) 11
*All questionnaires
**Including the removal of the sub-variable knowledge self-efficacy
The scales for the variable of SVO were not adjusted. These were tested in research of Murphy et al. (2011) as reliable and valid. Therefore, the researcher assumed that this is correct. The next step in the analysis is to construct the descriptive statistics and the correlation table in SPSS.
3. Descriptive statistics and correlations
As shortly introduced in the previous chapter, the sample is marked by equality around gender as well as SVO. In addition, most students are regular students who are on average 24 years old. Most of them study Business Studies and are in their Master program, mostly familiar with group assignments consisting of 4 people on average. The correlation table, including the descriptive statistics, is provided in table 6 below.
The correlations also show some interesting patterns. The knowledge sharing motivations for individual and group reward variables are positively correlated 0.231 with a significance level of P<0.01. This makes sense as both variables contain the same statements about knowledge sharing motivations with respect to a different type of reward (individual or group).
The table also shows that group size is negatively correlated to social value orientation (-0.178 with P<0.05). This means that whenever there is a larger group size, the proportion of individualists becomes higher. This is an interesting finding, which will be addressed in the discussion section.
Group size also shows a positive correlation with educational level (0.189 with P<0.05). This means that it is more likely that students on their Master level work in larger groups (except for the executive students). This is checked through examining the course guides of the faculty over the last five years. All the courses increase in the number of group assignments and in the size of the groups over the years from first year Bachelor, to third year Bachelor and Master.
4. ANOVA
The analysis for this thesis is done using a repeated measure ANOVA. The following assumptions for this type of ANOVA, following Tabacknick and Fidell (2001) and Field (2005), are checked and shortly discussed below.
-‐ The dependent variable needs to be continuous, which is the case as it is measured on a scale and averaged over several items.
-‐ The independent variable needs to consist of at least two levels. This is met as this variable includes both the group and individual reward.
-‐ The third assumption states that there need to be no significant multivariate outliers. This is tested using the Mahalonobis distance (Tabacknick and Fidell, 2001, p. 74 – 75). Here two outliers were found and removed from the dataset in order to meet this assumption.
-‐ The distribution of the dependent variable needs to be approximately normally distributed. This is checked through a normality test in SPSS, which shows that this assumption is met.
-‐ Normally there is also the ‘sphericity test’ that needs to show non-significance. However, this test is only for when there are three or more variables within the main independent variable (SPSS labels them as ‘within variables’) (Field, 2005, p. 428). It is not necessary to check this assumption, as the within variable in this research consists of two level (and not three or more).
-‐ The homogeneity of variance is tested using ‘Levine’s test’. The results need to be not significant. The results show that this assumption is met. The results are for the individual reward 0.099 and for the group reward 0.146.
-‐ The last assumption that is required is called the ‘Box M’ test, which tests the homogeneity of covariance. This needs to be non-significant. The results show a value of 0.226, which means that there is no significance found.
To conclude, all assumptions are met in order to perform a repeated measure ANOVA in SPSS. The results of the ANOVA are shown in table 6. In the table, the values for F, the significance levels (p) and the partial eta squared (η2) are given.
Table 6 – Results of the ANOVA
Model F p η2 Age 0.188 0.666 0.001 Gender 0.341 0.560 0.003 Education 0.048 0.826 0.000 Group size 1.745 0.189 0.013 Reward systems (RS) 5,341** 0.022 0.040 SVO 0.076 0.784 0.001 RS*SVO 0.568 0.452 0.004 **Significant for P<0.05
The first hypothesis stated that a student that receives a group reward is more likely to share knowledge. From the ANOVA the variable of reward system is shown to be significant, where F (1, 129) = 5.341, p = 0.022, η2 = 0.040 with p<0.05. The
graphical representation provided in SPSS is shown below in graph 1. The graph shows that there is a higher level of motivation to share knowledge whenever students receive a group reward (note that 1 stands for the individual reward and 2 for the group reward). Therefore, hypothesis 1 is supported.
Graph 1 – Model
The second hypothesis stated that prosocial students share more knowledge than individualists. The results are shown in model 2 and indicate that there is no significance found, with F (1, 129) = 0.076, p = 0.784, η2 = 0.001 with P<0.05.
Therefore, hypothesis 2 is not supported.
The third hypothesis is also shown in model 2. Here it is tested whether the relationship between reward systems and the motivation to share knowledge is altered by the SVO. Unfortunately the results show that there is no significance found with F (1, 129) = 0.568, p = 0.452, η2 = 0.004 with P<0.05. Therefore, the third hypothesis
is not supported.
The partial eta squared shows how much the independent variable explains the dependent variable by calculated a ratio based on the effect size divided by the effect size plus the associated error variance (Tabachnick and Fidell, 2001). Table 6 shows
the η2 of the between variables together only account for 1.8% of the model. The
variable reward systems accounts for 4.0%, which indicate that basing the reward on an individual or group level only plays a small role in the motivation to share knowledge. The interaction effect of reward systems and SVO only account for a small fraction of 0.4%. This shows that reward systems play a more important role than the interaction effect, however it is still a minor role in the entire model. In total, the variables in this model only account for 6.2%, which indicate that there are other factors that play a more important role in the motivation to share knowledge. In addition, these low values might also be an effect of the low sample size as this decreases the statistical power of the analysis.
The analysis shows some results that need further discussion. This will be provided in the next chapter.
6. Discussion
The main goal of this thesis is to analyze the effect of reward systems on the motivation to share knowledge, hereby taking into account that people differ in their SVO. In this model the first hypothesis stated that people tend to share more knowledge whenever they are rewarded as a group. In addition, the second hypothesis stated that prosocial students tend to share more knowledge. The conceptual model was expected to show that students that receive a group reward tend to share more knowledge, especially when that group consists of individualists. The results did show that a group reward increases the motivation to share knowledge. However, all other hypotheses were not supported. This following section provides a discussion of the results. Additional literature will be used in order to discuss the results. Furthermore, there will be an elaboration of the implications of the results for managers. At last the limitations and future research recommendations are given.
1. The effect of reward systems on knowledge sharing motivation
The results show that reward systems have an effect on the motivation to share knowledge. In addition the graph showed that students shared more knowledge whenever they were rewarded for the entire group. An explanation for this results could be the fact that everyone learned whenever sharing they shared their knowledge. This, according to the literature, results eventually in a higher grade (and thus a higher reward) for the one who shared and, so, for the entire group (Webb, 1982). The data of this research shows the same results; students received on average a 7.71 for a group assignment compared to a 7.45 for an individual assignment.
Even so, the partial eta squared showed that the two levels of reward systems only account for 4% in the model. This means that there are other factors that play a
more important role. An example is that reward systems consist of more factors besides the level of reward. This is included more detailed in the future research recommendations.
2. The effect of SVO
The analysis does not provide proof that SVO is a moderator on the relationship between reward systems and knowledge sharing behavior. The literature suggests some form of relationship. For example the research of Galletta et al (2003), they found strong significant results for the direct relationship between SVO and the motivation to share knowledge. They measured this relationship conducting an experiment among 76 students. The difference with the research here is that it analyses a moderating effect. Even so, analyzing the direct effect between SVO and the motivation to share does not lead to significant results in the data (the value F (17, 144) = 1.575, p = 0.080).
The research examined students that both received a group and individual reward for a course. However, it did not look at the distribution of students in a group. The correlation showed that there is a negative relationship between group size and social value orientation. However, group size did not influence knowledge sharing motivations. So, the proportion of individualists in a group might not have an effect on the behavior of that group. Following Bridoux et al (2011), as long as there are reciprocators in a group, whenever there is a fair treatment, it will encourage other group members to also reciprocate. Here there is indication that this might be the case. This implicates that even though there is a SVO, the individual’s behavior could be influenced by the behavior of reciprocators in the group. In turn, the focal individual’s behavior motivates (or perhaps also demotivates) people to contribute to a group.
The Murphy slider (Murphy et al, 2011) in this thesis is used to indicate the ‘first degree measure’. The research of Murphy et al (2011) also indicates that there are different motivations for prosocial people to reciprocate. This is called the ‘second degree measure’. An individual is prosocial whenever they focus on benefitting the entire group, so whenever everyone receives an equal amount of money. Another motivation is that this person has an ‘inequality aversion’, here this person has a preference of equality within a group. Prosocial individuals tend to reciprocate, but apparently there exists two reasons why someone is in fact prosocial; either the same benefit for each member of the group or in order to assure equality within the group. In order to analyze which motivation a prosocial has in the sample of this research, additional items were asked. Here in total out of all prosocial people, only four were indicated as ‘joint gain-types’ and sixty-four were acknowledged as ‘inequality aversion’. This supports the argument in the previous paragraph that the group might encourage and/or demotivates group members to engage in sharing knowledge. If the preference is to avoid inequality, than that group member is prone to share knowledge when there is a risk of inequality, and as soon as there are reciprocating individuals, not reciprocating would increase inequality.
3. Additional results
The analysis found some additional results. The literature review states that students with a higher educational level usually share their knowledge more often (Bruffee, 1995). Students on a higher educational level are often older, as they have to go through several educational levels. The correlation table showed a correlation between group size and education. Even so, there was no correlation found between age and group size, or between education, age, group size and knowledge sharing motivations.
4. Managerial implications
This thesis has important managerial implications. Perhaps the most important one is that people are (de)motivated to sharing their knowledge with others whenever there is a specific reward system (here it is either group or individual), while also considering the fact that they perceive behavior of others differently through their SVO. For a manager it is important to consider this whenever setting up a reward system, as both that reward and the SVO causes a reaction by their employees. Especially in firms that rely on knowledge intensively it is important to state how to organize such a firm in order to encourage an easy transfer of knowledge.
Knowledge transfer is such an important topic that it has kept scholars busy for a long period (Goh, 2002; Riege, 2005). It is important because it is argued to have enormous benefits for an organization. Most papers consider either micro level research or macro level research, where the aim is to get insight into how to organize the organization on the one hand. On the other hand the aim is also how they need to manage people. This thesis has provided insight into the link between these two, which provides a better understanding for managers. The insight that reward systems
already encourages people to share knowledge is perhaps also showing the potential benefit that knowledge sharing can have on in an organization (Riege, 2005).
5. Limitations and future research recommendations
The research performed for this thesis has some limitations and, therefore, also some future research recommendations. The first limitation is the relatively small sample size (of 149 respondents). The primary sample was a little over 200, however due to missing values it was diminished with 25%. Out of a total population of almost 9,000 people the sample would ideally be around 360 respondents. In addition, (too) small sample sizes are prone to insignificant results due to this small size (Cohen, 1962).
Another limitation, which might also explain the low respondents rate, is the fact some research argues that economists and business people are usually more self-interested (Carter and Irons, 1991; Frank, Gilovich and Regan, 1993). This is a limitation, as the entire sample only consists of economists and business students. Even so, economists are also seen as cooperative, which would argue that the results are still to some extend generalizable (Frank, Gilovich and Regan, 1993). Especially considering the fact that this sample showed an even amount of individualists as well as reciprocators. It would be interesting to extend this research by including all students from the University in Amsterdam in other to analyze whether the results are indeed generalizable. Another recommendation could be to, in order to test the research question, approach all employees at the university. This, perhaps, approaches the organizational ambiance better than to focus solely on students.
There is an indication that the results are nonetheless generalizable, because the students that responded to the survey will start working in the near future. Even so, the effect on people of receiving a grade as opposed to receiving a monetary reward
differs. There are several research articles that use a different approach for reward systems, namely monetary and non-monetary (Bartol and Srivastava, 2002; Lin, 2007; Foss et al, 2010). This is a limitation of this study, as students only obtain a non-monetary reward. Future research should therefore include these types of rewards in order to see if this causes differences in behavior.
Another future recommendation is to conduct the same research using an experiment following Galletta et al (2003). Using an experiment diminishes the level of self-bias. Here the researcher is able to observe the behavior of the respondents (Saunder, Lewis and Thornhill, 2011), where in a survey the respondent fills in what he or she believes will do.
On a more meta-level it is advisable to extend research on the individual motive to share knowledge, apart from the SVO (Bock and Kim, unknown year; Riege, 2005). There is potential in understanding why people want to share what they know and especially why they refuse to do so. In order to analyze this, I would advice to perform a multilevel analysis including several personal traits.
7. Conclusion
In this thesis the role of both reward systems and SVO on knowledge sharing behavior among students was investigated. Here, the SVO of students, e.g. how they perceive behavior of their co-students, is examined as a moderating effect on the relationship between reward systems and knowledge sharing behavior. This model was tested using a two-way ANOVA, as the reward systems consisted of both an individual and a group base.
The results show that students share more knowledge whenever they obtain an individual grade, regardless whether they work individually or in a group. Unfortunately, there is no relationship found between SVO and knowledge sharing behavior. In addition, the conceptual model was tested as not significant. Future scholars are still advised to continue research in this specific area as the literature suggests that this model exists in reality.
This research showed that reward systems have an effect on knowledge sharing behavior. It is important for managers to consider whether they reward their workers individually or on a group, depending on the desire of the manager and the organization. Focusing on how to organize the organization while taken into account that each person behaves differently towards outcomes contributes to perhaps competitive advantage. Future research should focus on how micro- and macro-level operations within the firm can facility an easier transfer of knowledge.
8. Reference list
Anderson, J. R. (2005). The Relationship Between Student Perceptions of Team Dynamics and Simulation Game Outcomes: An Individual-Level Analysis. Journal of Education for Business, 81(2), 85–90.
Augier, M., Shariq, S. Z., & Vendelø, M. T. (2001). Understanding context: its emergence, transformation and role in tacit knowledge sharing. Journal of Knowledge Management, 5(2), 125–137.
Baaij, Marc. An Introduction to Management Consultancy. Sage, 2013, first edition. Barney, J. B. (1986). Strategic factor markets: Expectations, luck, and business
strategy. Management science, 32(10), 1231-1241.
Barney, J.B. (1991). Firm resources and sustained competitive advantage. Journal of Management, 17(1), 99 - 120.
Bartol, K. M., & Srivastava, A. (2002). Encouraging Knowledge Sharing: The Role of Organizational Reward Systems. Journal of Leadership and Organizational Studies, 9(1), 64–76.
Biemann, C. (2006). Chinese whispers: an efficient graph clustering algorithm and its application to natural language processing problems. In Proceedings of the first workshop on graph based methods for natural language processing (pp. 73–80). Association for Computational Linguistics. Retrieved from
http://dl.acm.org/citation.cfm?id=1654774.
Bock, G. and Kim, Y. (unknown). Breaking the myths of rewards: an exploratory study of attitudes about knowledge sharing. Not yet published, received via gwbock@gmail.com.
Bock, G., Zmud, R., Kim. Y. & Lee, J. (2005). Behavioral intention formation in knowledge sharing: examining the roles of extrinsic motivators,
