Abstract
Creativity and team work are important factors in the current economy. Research has shown that team level creativity depends on individual level creativity, but having creative team members may not guarantee creative teams. Some evidence suggests that cooperation moderates the relationship between individual and team creativity. A high level of
cooperation can be achieved by implementing a reward structure based on team level rewards. In this study we argue that reward structure (individual vs. team based) affects the level of cooperation in teams, and that cooperation has a positive effect on the relation between individual and team creativity. We studied this in a lab study with 40 groups consisting of 3 respondents performing individual and group creative tasks, and in which half the groups were rewarded based on individual level performance and the other half of the groups on group level. It was found that individual creativity has a positive relation with team creativity. However we did not found a moderating effect of cooperation on this relationship. We did found that it is a better idea to reward teams based on team performance instead of individual performance, especially when teams consist of creative and uncreative members.
Keywords: Individual creativity, team creativity, cooperation, reward structure, team
In our current world, creativity and team work are two important factors. Creativity is needed for companies to be and stay innovative, and by being innovative, reaching a competitive advantage compared to competitors. All innovation begins with creative ideas of individuals; these creative ideas represent individual creativity (Amabile, Conti, Coon, Lazenby & Herron, 1996). Team work is also very important in the current economy. Facts show that in 1996 already 66% of the large firms had 20% or more employees working in teams. Team work is also important because teams can be more productive in certain situations than individuals (Lazear & Shaw, 2007). When the quality of team work is high, the output of the team and the company will be high. Although we know much about individual creativity and team work, research has only recently began to address the issue of creativity in teams.
creativity. Furthermore, we want to study if diversity in individual level creativity in a group affects team creativity, and whether this relation also depends on cooperation. In the current literature, the average of individual level creativity is often used in studying the relationship between individual and team creativity (Miron-Spektor et al. 2011, Pirola-Merlo & Mann, 2004, Taggar 2002). We extend this work, by examining the relation between diversity in individual-level creativity and team creativity.
An important factor influencing cooperation is reward structure, and especially
whether group members are rewarded at the individual or at the group level. Cooperation will be higher when group rather than individual performance is rewarded, and evidence suggests that group creativity benefits more from group than from individual rewards (Bechtoldt, de Dreu, Nijstad & Choi, 2010). Because reward structure affects cooperation, and cooperation is expected to moderate the relation between individual and group creativity, we expect that the relation between individual-level creativity and group creativity is stronger when groups are rewarded for group rather than individual performance. This prediction is tested in a
laboratory study, in which individual and group creativity are assessed, and in which reward structure is manipulated.
Theoretical Background
Individual creativity and Team creativity
Creativity can be defined as a judgment of the combined novelty and usefulness of something (Pirola-Merlo & Mann, 2004). In this study we examine whether and when individual level creativity has a relationship with team level creativity. According to Taggar (2002), individual creativity accounts to a large extent for team creativity. Individual creativity consists to a large extent of raw materials provided by the individuals. These raw materials are
on these three predictors individual creativity is likely to be high as well. He also states that individual creativity is positively related to team creativity however he also mentions that this relation is moderated by several variables.
There is a second reason why individual and team creativity will be positively related. According to Zhou (2003), the presence of highly creative team members will increase the creativity of other members, and especially for the least creative members, this leads to higher creativity. Choi & Thompson (2005) found that individual creativity could lead to team creativity by means of newcomers. When newcomers with high creativity enter the group the team creativity increases as well because of the new point of view of the newcomer. These new points of view did also influence the creativity levels of members who were in the group for a longer period. This evidence proves that individual creativity of member does have a positive effect on the creativity as a team. Another effect was found by Miron-Spektor et al. (2011). They found that teams consisting of creative members scored higher on radical team creativity. This means that a team with highly creative members is more likely to make big creative changes, because they have the creative capacity from the highly creative individuals in their team. This evidence showed that individual creativity does have a positive effect on team creativity. In this study we expect to find the same relationship. Our first hypothesis is:
H1: Individual creativity (the average creativity of team members) has a positive effect on team creativity
Reward structure, cooperation and how this moderates the relationship between individual and team creativity
other members’ ideas, and this critically depends on within-team cooperation (see e.g., Bechtoldt et al., 2010). Indeed, some evidence indicates that individual creativity is more strongly related to team creativity when the level of cooperation in the team is high (Taggar, 2002). In particular, Taggar (2002) has found “group relevant processes” stimulate the effect between individual level creativity and team level creativity. These group relevant processes are: inspirational motivation, organization and coordination and individual consideration. These effects indirectly represent cooperation, which we use in our model. Vegt van der & Janssen (2002) found that in heterogeneous groups task interdependence was strongly and positively related to creative behavior of individuals with high levels of goal interdependence. This means that when team have a high level of interdependence the team creativity will be influenced positively.
One way to stimulate cooperation is by rewarding collective team performance, rather than individual performance. Indeed, research has shown that reward structure has a large influence on the cooperation in a team. When rewards are based on equality cooperation will be higher than when rewards are based on equity (Chen, Chen & Meindl, 1998). This means that cooperation will be higher when each team member receives an equal part of the reward and will be lower when each person gets a reward that is proportioned to their contribution. Another evidence of the importance of team based rewards is provided by DeMatteo, Eby & Sundstrom (1998). They also found evidence that team based rewards are fostering
team, they have more motivation to work with their team to achieve the reward. When team members are rewarded as individuals they are less motivated to work with their team, and more focused on achieving their individual reward. This is why we state the following hypothesis:
H2: Team reward structure has a positive effect on cooperation.
Individual creativity as a basis is not enough to create the highest team creativity, and according to Taggar (2002) the way of interaction is also very important to achieve the highest team creativity. Creativity at an individual level may not always ensure creativity of the group as a whole. Taggar (2012) also states that in achieving team creativity the
effectiveness and cooperativeness of group processes are very important. This cooperation makes it possible for individual team members to exchange their creative ideas in an effective way, and by thereby creating more team creativity.
a team is well coordinated and organized, and when they evaluate ideas well the creativity of the individual team members will create more team creativity. The coordination and
cooperation is important because the quality of ideas must be recognized to create the most team creativity. This means that cooperation as mentioned by Brophy (1998) moderates the relationship between individual and team creativity positively. In our study we expect, as based on the literature provided above, to find a positive moderating effect from coordination on the relation between individual creativity and team creativity. This results in the following hypotheses:
H3: Team-based as compared to individual rewards strengthen the relationship between individual creativity and team creativity
In sum, theoretical arguments and empirical evidence suggest that individual creativity has a positive effect on team creativity (e.g., Miron-Spektor et al., 2011; Pirola-Merlo & Mann, 2004). Furthermore, team reward structure will have a positive effect on cooperation (Bechtoldt et al., 2010, Chen et al., 1998), and cooperation strengthens the relation between individual and team creativity (e.g., 2002). This suggests that the moderating effect of reward structure on the relation between individual and team creativity is mediated by cooperation. Our final hypothesis therefore is:
H4: The moderating effect of reward structure on the relation between individual and team creativity is mediated by cooperation
Figure 1. Conceptual model
Diversity
In this experiment we will try to contribute to literature in the case of what kind of relation diversity at individual level has on team creativity. We found some evidence that this relation is present. Van der Vegt & Janssen (2002) state that heterogeneous groups perform better on innovation than homogeneous groups. This because a diverse team would be able to pool different ideas and combine this information to higher innovativeness. Miron-Spektor et al. (2011) state that if a team consists of highly creative persons strong radical innovation
interdependence high, cooperation will also increase. Diversity could have a positive effect on cooperation, which makes it interesting to incorporate diversity in our experiment.
Method
Participants and Design
To test our hypotheses we performed a lab study. In this study we formed groups of three team members, and a total of 40 groups, with 120 respondents in our study. Of these 120 persons, 49.2% was male and 50.8% female, with an average age of 22.02 and a standard deviation of 2.25. Of respondents, 60.0% was Dutch, 10.8% was German, 6.7% Bulgarian and the other 22.5% of the persons are from 18 different countries. In our study there were two conditions: in the first condition, the three best team members could receive a €50. – reward, based on the best individual performance; in the second condition the best overall team could receive a €150, based in the best team performance. The first condition was tested on 20 groups and the second condition was tested on the other 20 groups.
The second independent variable was individual creativity. This was measured in three different way. First, we used the scale from Gough (1979), which measures several
innovative or creative the person is. Respondents have to rate themselves on statements such as “I have original ideas “. We tested KAI on reliability and Cronbach’s Alpha was .79, which makes it a reliable scale to use. Individual creativity was also measured by using a
brainstorming session, in which individual respondent had to come up with as many answers as possible in 5 minutes. The more answers the respondents produced, the more creative they were.
Individual creativity scores were aggregated to the group level in two different ways. First, we averaged individual creativity for each group, to examine whether groups with more creative individuals on average were more creative. Second, we established the standard deviation of individual creativity within the groups, to examine whether diversity in creativity of members is associated with creativity of groups.
Procedure
After the respondents came in to the research lab, and completed an informed consent procedure, they were tested individually on their creativity level, by filling in the Gough scale and the KAI (Kirton) scale. We then measured creativity in a third way, by having
respondents perform an individual brainstorming task, in which they had to write down as many original answers as possible in 5 minutes about ways to use a brick. This was a paper-pencil test and all respondents were in the same room together sitting at different tables were they could not see ideas of others.
possible about ways to use a football. They were asked to write down their own ideas on an individual piece of paper that was divided in boxes, and were asked to write one idea per box. After the brainstorming session, participants individually filled in a questionnaire, in which we measured the level of cooperation in their team during the brainstorming session. After that they received a debriefing about the purpose of the research and that ended the
experiment.
Dependent Variables
After the team brainstorming session cooperation in the group was measured. We measured this by using a self-developed cooperation scale. This scale measures the
cooperation perceived by the team members during the team assignment with statements such as: “Every member was able to finish their contribution without interruption from other members.” The respondents have to respond to the statements by using a 5 point likert-scale on 13 items. To test if this scale is reliable we did a reliability test, which resulted in a Cronbach’s Alpha of 0.87. This makes the scale reliable to use.
know when you cannot park any further with you car. When these score of the ideas are summed up and divided by the number of ideas we have a new measure called average creative score. Ideas with a score of three were counted for each group to arrive at a measure of total number of good ideas. To make sure that the idea ratings were reliable we used a peer assessment. All ideas were rated by a first rater, and a second rater unaware of study
hypotheses also rated 120 ideas based on the scale presented above. This peer assessment resulted in a Cohens Kappa of 0.925 which indicates that ratings were reliable.
Results
In our analysis we have three measures that represent individual-level creativity namely: Gough score, Kirton score, and Number ideas individual brainstorming session. Together with the manipulation of reward structure, these are the independent variables. For team creativity we have three measures namely: number of ideas in the team brainstorming session, average score of ideas in the team brainstorming session, and number of good ideas in the
brainstorming session. These are the dependent variables.
Descriptive statistics
We started our analysis with descriptive statistics and correlation. We did this at individual level (Table 1) and at team level (Table 2). At individual level (Table 1) gender was
there was a marginally significant negative correlation between number of ideas and the average creativity of those ideas, and a significant correlation between average creativity and number of good ideas (Table 2).
Table 1: Descriptive statistics individual level.
Note: + p < .10; * p < .05; ** p < .01; *** p < .001
Table 2: Descriptive statistics Team level.
Note: + p < .10; * p < .05; ** p < .01; *** p < .001
Mean SD Gender Age Gough Kirton Number
Ideas Gender 0.51 0.50 - Age 22.02 2.25 .03 - Gough 4.58 2.94 -.33*** -.06 - Kirton 2.85 0.32 -.21* -.05 .32*** - Number Ideas 9.98 3.97 .06 .09 .09 .20* -
Hypothesis Testing
To test the hypotheses, the three dependent variables (number of ideas, average creativity of ideas, and number of good ideas) were analyzed by means of regression analysis. We created a dummy variable for condition, in which the individual reward condition was coded with 0 and the team reward condition with 1. Individual level creativity was measured in three ways (Gough and Kirton scales, and mean number of ideas in the individual brainstorming session) and these three variables were aggregated to the team level in two different ways (taking the mean and standard deviation), creating a total of six different measures. These six measures were z-transformed before analysis. Further, these z-scores were multiplied with the dummy for condition, to create interaction terms and examine interactions between individual-level creativity measures and reward structure (a total of six interaction terms).
We then performed a total number of 18 regressions, in which we examined the effects of the six measures of individual creativity, together with condition and the interaction
between condition and the individual creativity measure, on the three measures of team creativity. No effects were found in analyses with the Gough scale, and no effects were found on the number of good ideas, and these results are further ignored. However, some effects were observed for both the Kirton scale (team mean and standard deviation) and for
Table 3: Results regression
Effects on average quality team creativity
Hypothesis 1 predicted that individual creativity would be positively related to team creativity. Results with regard to this hypothesis are mixed. On the team number of ideas, there is a significant positive effect of mean member number of ideas, and a non-significant positive effect of the mean Kirton score. However, on team average idea creativity, both mean member number of ideas and mean Kirton score have significant negative effects. In all, Hypothesis 1 should therefore only be accepted for number of ideas but not for the average quality of ideas. This effect appears because team member were asked to think of as many ideas as possible instead of focusing on the quality of these ideas.
Hypothesis 3 predicted that the relation between individual and team creativity would be stronger with a team rather than an individual level reward. On team number of ideas, the relevant interactions were not significant. However, on team average creativity, the
Independent Variables Dependent variables Number of Ideas (team) Mean score creativity (team) Condition b(t)= -1.28(-0.40) b(t)= -0.01(-0.24) Number of ideas (in.) Mean b(t)= 6.57(2.78)* b(t)= -0.06(-2.70)**
Interaction b(t)= 1.97(0.60) b(t)= 0.05(1.85)+
R² 0.38 0.17
Condition b(t)= -0.93(-0.24) b(t)= -0.01(-0.24) Number of ideas (ind.) Standard deviation b(t)= -2.18(-0.78) b(t)= -0.01(-0.43) Interaction b(t)= 7.75(1.97)* b(t)= 0.00(0.07) R² 0.11 0.01 Condition b(t)= -1.65(-0.43) b(t)= -0.01(-0.24) Kirton Mean b(t)= 4.23(1.64) b(t)= -0.05(-2.38)* Interaction b(t)= 0.02(0.01) b(t)= 0.07(2.27)* R² 0.12 0.16 Condition b(t)= -1.03(-0.29) b(t)= -0,01(-0.17) Kirton Standard deviation b(t)= -1.77(-0.65) b(t)= -0.04(-1.87)+ Interaction b(t)= 9.87 (2,72)** b(t)= 0.03(0.98)
interaction between mean number of individual ideas and rewards was marginally significant, and the interaction between mean Kirton score and rewards was significant. Figure 2 shows that the relation between mean number of individual ideas and average team creativity is negative in the individual reward condition, but that there is no relation between these variables in the team reward condition.
Figure 2: Individual number of ideas in relation to average score of ideas team level (condition)
Similar results were obtained with the mean Kirton score as a measure of individual
creativity. As shown in figure 3, in the individual reward condition, the relation between mean Kirton score and team average creativity is negative, whereas this relation is weaker and positive in the team reward condition. Based on the evidence from table 3 and Figure 2 and 3 we can conclude that we have to reject H3. Although there is some evidence that rewards moderated the individual-team creativity relation, this relation is not stronger and positive in
0,98 1 1,02 1,04 1,06 1,08 1,1 1,12 1,14 1,16 1,18
Low Mean_number High Mean_number
the team reward condition. Rather, this relation becomes negative in the individual reward condition, and this only applies to a measure of average team idea creativity.
Figure 3: Kirton score related to average score ideas team
Cooperation and rewards
To test the relation between reward structure and cooperation (H2) we did a independent sample t-test, with reward condition as independent variable and cooperation as dependent variable. Although cooperation was slightly higher in the team reward condition than in the individual reward condition, the difference was not significant, t (38)= -1.32, p = .20, and hypothesis 2 should be rejected. Because we had to reject the second and third hypothesis we also have to reject the fourth hypothesis which assumed a relation between rewards and cooperation and a moderation of the individual-team creativity relation by rewards.
1 1,02 1,04 1,06 1,08 1,1 1,12 1,14 1,16
Low Mean_kirton High Mean_kirton
Exploratory results
In this research we want to contribute to the literature by also analyzing creativity by means of within-team variance. We thus analyzed the effects on team creativity of the variance in number of ideas individually, and the variance in Kirton score. Results can be found in Table 3.
On average number of ideas, only the Kirton within-team standard deviation had a marginally significant negative effect. On the team number of ideas, however, there were significant interactions between the team standard deviation in individual number of ideas and condition, and between the Kirton within-team standard deviation and condition. Figure 4 shows that the standard deviation in number of individual ideas has a positive relation with team number of ideas in the team reward condition, but not in the individual reward condition. A similar effect was observed for within-team standard deviation in Kirton scores: this
measure is positively related to team number of ideas only in the team reward condition, but not in the individual reward condition (see Figure 5)
Figure 4: Number of ideas individual SD related to number of ideas team. 0 5 10 15 20 25 30 35 40
Low SD_number High SD_number
Figure 5: Kirton variance related to number of ideas team level
Discussion
Goals and method
In this research, we analyzed the relationship between individual creativity and team
creativity. We proposed that individual creativity is positively related to team creativity, but that this relation is moderated by reward structure, because team-based rather than individual rewards stimulate higher levels of cooperation. We further explored the effects of diversity in terms of individual creativity within teams. We performed a lab experiment in which we manipulated reward structure (group vs. individual rewards) and measured individual
creativity in various ways: By means of questionnaires (Gough and Kirton) and an individual 0 5 10 15 20 25 30 35 40 45
Low SD_Kirton High SD_Kirton
brainstorm session. We measured team creativity in a team brainstorming session, and cooperation by means of a questionnaire.
Summary of results
We tested our results in multiple ways and found that the Gough scale and average quality of ideas at individual level had no effects on group creativity, and we therefore focused on two other measures of individual creativity: number of ideas generated in the individual
brainstorm (individual fluency) and score on the Kirton Adapter Innovator Inventory (KAI). First, we tested whether individual creativity (the average creativity of individual team members) is related to group creativity, and whether this relation is moderated by reward structure by means of a regression analysis. The results from this test showed some evidence for a positive relationship between individual and team creativity: on the number of ideas generated in the group brainstorm, there was a significant positive effect of individual fluency and a non-significant positive effect of the KAI. However, on average quality of ideas,
negative effects of both individual fluency and KAI were observed. Contrary to hypotheses, we further found a negative significant moderating effect of team reward structure on the relation between individual creativity and average quality of ideas (team creativity). We found no relation between individual and team creativity when rewarded as a group. However when rewarded as a individual in the team assignment there is a strong negative relation.
To explain this pattern of results, it is important to note that in both reward conditions it was explicitly stated that rewards would be determined based on the number of ideas generated, rather than their quality. It may thus be the case that creative individuals focused on generating many rather than high quality ideas. Furthermore, the number of ideas
individual creativity on average idea quality only occurred in the condition with individual rewards is potentially due to the fact that especially very creative persons are not likely to share creative ideas when others benefit from their ideas as well and they might lose the competition by helping others.
By means of a t-test we tested if reward structure had a positive effect on the
cooperation within the team. Here we did not find a significant effect, although cooperation was slightly higher in the group reward condition than in the individual reward condition. This means that reward structure has no effect on the cooperation in this experiment, which is potentially due to relatively high scores on cooperation across conditions (i.e., a ceiling effect). For our model this mean that the moderating effect of rewards structure on the relationship between individual and team creativity cannot be explained by cooperation.
The second part of our research consisted of an exploratory study of the effect of variance in individual level creativity in relation to team creativity. Our main findings are that variance at individual level is positively related with team creativity, but only when rewarded at the group level and not in the individual reward condition, and only for the number of ideas generated in the group. We can explain this effect in the way that when there is high variance within a group, highly creative member(s) would let the others benefit more from that
Implications
Theoretical implications
Pirola-Merlo and Mann (2004) found that team creativity can be described as the sum of creativity of the individuals in the team. Our results suggest that this is not always the case. Indeed, we even found a negative relation between individual creativity and average quality of ideas (team creativity) in the individual reward condition. This means that team creativity is not just the sum of individual creativity, especially when the team are rewarded at
individual level. For a individuals to be creative in a team setting we propose a group reward instead of a team reward, especially when there is high mean individual creativity or a high variance in individual creativity. Taggar (2012) stated that cooperation would moderate the effect between individual and team creativity. Unless we did not found this relation (nearly significant) we do think that because of group rewards people will cooperate better which then makes them more creative as a team.
Bechtoldt et al. (2010) found that group creativity benefits more from group rewards than from individual rewards. We have to nuance this statement because in our research we found a positive effect from group rewards on creativity, but only when there is high variance of individual creativity within a group. This means that Bechtoldt et al. (2010) are right but only under the condition that there is a high average creativity within a group or high variance of individual creativity within a group.
Practical implication
Individual rewards seem to undermine the performance of teams with highly creative member. Team rewards are a better idea especially when teams consist of a mixture of creative and uncreative persons. In that case the most creative person will pull the creativity levels of the others to a higher level, resulting in a better overall performance.
Limitations and future directions
We did find some interesting results in this experiment, but there are some limitations to mention. We used 120 respondent in a lab setting. This makes it hard to generalize the findings from this experiment. It might be better to repeat the experiment in a more natural setting and with more respondents, which would make the results more generalizable.
We did not find an effect of reward structure on cooperation in our experiment. This could be due to the short time period people had to cooperate with each other. In our results we found very high scores of the cooperation scale which might point to a socially desirable answer score given by the respondents. In a future experiment it might be good to let the respondents work together longer or on a more difficult task in which they have to cooperate more intensively. This might get the respondents to the points in which they can judge their cooperation in a more reliable way.
We found some relations between individual and team creativity. In the results section we mention two predictors for individual creativity and two for team creativity. However there are more variable mentioned at the beginning of the results section. The other variables were supposed to measure individual creativity or team creativity did not show any significant results. This means that the results should be treated with some caution.
For future research we suggest to look for other moderators in the relationship
could be done by rewarding one person in each team, based on his or her performance, so individuals within the team compete for the reward. On the one hand, this would be a strong motivator for team members to individually perform well, but on the other hand may
negatively affect team creativity because it undermines the cooperative sharing of ideas.
Conclusion
In this experiment we examined if there is a positive relation between individual creativity and team creativity, and whether this relation is moderated by cooperation and reward structure. We found some evidence that there is a positive relation between individual and team creativity when creativity was measured as the number of ideas produced in a
brainstorming session. However, when team members were rewarded for their individual performance, we found a negative relation between individual and team creativity measured as the average creativity of ideas. It thus appears that generating many ideas decreased the average creativity of ideas, but only in the individual reward condition. We finally found that teams that have a large variance in individual level creativity scored higher on team level creativity (number of ideas) in a group reward setting then teams with less variance. In sum, to reap the benefits of individual creativity in teams, it is better to reward performance of teams rather than individuals.
Literature
Amabile, T.M. 1983. The Social-Psychology of Creativity – A Componential
Conceptualization. Journal of Personality and Social Psychology, 45(2), 357-376
Amabile, T.M., Conti, R., Coon, H., Lazenby, J. & Herron, M. 1996. Assessing the work environment for creativity. The Academy of Management Journal, 39(5), 1154-1184.
Bechtoldt, M.N., Dreu de, C.K.W., Nijstad, B.A. & Choi, H. 2010.Motivated information processing, social tuning, and group creativity. Journal of Personality and Social Psychology, 99(4), 622-637.
Brophy, D.R. 1998. Understanding, measuring, and enhancing collective creative problem- solving efforts. Creativity Research Journal, 11(3), 199-229.
Chen, C.C., Chen, X.P. & Meindl, J.R. 1998.How can cooperation be fostered? The cultural effects of individualism-collectivism. Academy of Management Review, 23(2), 285-304.
DeMatteo, J.S., Eby, L.T. & Sundstrom E. 1998. Team-based rewards: current empirical Evidence and direction of future research. Research in Organizational Behavior, 20, 141-183
Dreu de, C.K.W., Weingart, L.R. & Kwon, S., 2000. Influence of social motives on
integrative negotiation: A meta-analytic review and test of two theories. Journal of Personality and Social Psychology, 78(5), 889-905.
Gough, H.G. 1979. A Creative Personality Scale for the Adjective Check List. Journal of Personality and Social Psychology, 37, 1398-1405.
Kirton, M. 1976. Kirton Adaption- Innovation Inventory (KAI). Journal of Applied Psychology, 61, 622-629.
Knippenberg van, D., & Schippers, M.C. 2007. Work group diversity. Annual Review of Psychology, 58, 515-541.
Lazear, E.P. & Shaw, K.L. 2007. Personnel Economics: The Economist’s
View of Human Resources. Journal of Economic Perspectives, 21(4), 91-114.
Pirola-Merlo, A., & Mann, L. 2004. The relationship between individual creativity and team creativity: Aggregating across people and time. Journal of Organizational Behavior, 25, 235-257.
Taggar, S. 2002. Individual creativity and group ability to utilize individual creative resources: a multilevel model. Academy of Management Journal, 45, 315-330
Vegt van der, G.S. & Janssen, O. 2002. Joint impact of interdependence and Group diversity on innovation. Journal of Management, 29, 729-751
Vegt van der, G.S. & Vliert van der, E. 2005. Effects of perceived skill dissimilarity and task interdependence on helping in work teams. Journal of Management, 31(1), 73-89
