Abstract
Innovation, a process fueled by creativity, is key to organizational survival. The current studies test a multilevel moderated mediation model to explore whether team behavioral integration influences individual creativity in general management teams. Two field surveys were conducted: Study 1 included 356 employees nested in 86 teams; Study 2 included 138 employees nested in 39 teams. Results from integrated path analyses demonstrate that team behavioral integration is positively related to individual creativity, explorative and exploitative learning mediate the relationship, and the indirect effects are stronger for individuals with higher creative self-efficacy. Implications and suggestions for future research are discussed.
Keywords
As globalization and technology amplify competition within organizations, the future of work and the role of employees are continuously evolving. The key to survival in this ambiguous and volatile world is to constantly adapt and innovate. Yet innovation cannot occur in isolation or by will alone, given that it is largely driven by creative individuals and teams who are able to work together and utilize their strengths. Creative employees drive their organization’s innovative spirit by not only affecting individual work performance, but by also influencing team and organizational performance. An organization’s survival may therefore hinge on its employees’ ability to express their creativity, defined as the creation process of novel and useful new ideas, while taking into consideration the complex interaction processes and creative mechanisms that occur within diverse teams (Amabile, 1996; Woodman et al., 1993).
Additionally, in an effort to remain competitive in complex global market environments, organizations are relying on teams and team-based decision making (Li et al., 2010). A team composed of employees with different strengths, abilities, expertise, and relevant experiences can help improve the quality of the work produced and decisions made, especially for complex problems that require creative solutions (Carmeli & Shteigman, 2010). The collision of different perspectives stimulates individuals’ divergent and flexible thinking style, and improves individual creativity (Hutzschenreuter & Horstkotte, 2013). Therefore, paying more attention to the behavioral interaction among team members is helpful to better understand the antecedents and mechanisms of individual creativity improvement from a behavioral perspective.
Team Behavioral Integration
The degree of convergence of team member interactions, a concept labeled as top management team (TMT) behavioral integration, demonstrates the team’s collective behavior toward positive organizational outcomes (Hambrick, 1987). TMT behavioral integration has been linked with greater levels of decision making, organizational learning, and innovation (Wu et al., 2002), as those teams work at the highest level of the organization and influence the strategy and innovation processes (Papadakis & Barwise, 2002). While research on behavioral integration of top management teams is relatively mature (Hambrick, 1987; Tekleab et al., 2016), scant empirical evidence supports whether behavioral integration is as instrumental for general management teams, such as daily operations teams, and knowledge management teams (Qu & Liu, 2017; Tao, 2011; Xu, 2013). General management teams are involved in the executional processes of an organization (Awino & Bwire, 2018), engaging in day-to-day operations and creative endeavors. We assert that empowering general management teams to enact integration behaviors is related to employee learning, which influences creative process engagement (Alsharo et al., 2017). More specifically, the current study examines self-regulatory mechanisms that influence individual creativity in this process (Magpili & Pazos, 2018; Van der Haar et al., 2017).
Existing research on the influence of behavioral integration mainly focuses on the organizational or team level, while there are relatively few studies looking at the individual level as a dependent variable construct of focus (Liu & Jiang, 2015). Social learning theory (SLT; Bandura, 1977) provides a logical framework for the current study, given that it discusses the influence of individual cognition, behavior and situational factors and their interactions on human behavior, and asserts that individual creativity is the product of the interaction between individuals and their surroundings. Team behaviors, when framed as a situation factor, are proven to affect individual creativity (Somech & Drach-Zahavy, 2013). Social learning theory (Bandura, 1977) also posits that behavioral acquisition has two different processes: the process of acquiring behavioral response patterns through direct experience, and the process of learning behavior by observing through indirect experience. Based on the idea of vocationally-oriented learning for employees in organizations, team behavioral integration provides an important source of information for individual learning behavior. Organizational learning can have a tremendous benefit to enhancing both collaboration and individuals’ creativity (Zhang & Yu, 2015). Therefore, the current research introduces organizational learning as a mediator through which team behavioral integration is related to individual creativity.
The present research also focuses on creative self-efficacy, defined as an individual’s belief in their ability to produce creative outcomes, because it has been linked to creative performance both at the individual and team level over time (Tierney & Farmer, 2002, 2011). This is vital to general team performance because engagement in learning also generates goal interdependence, and strong internal and external communication (Pirola-Merlo & Mann, 2004). This study applies the concept of top management team behavioral integration to general management teams, and emphasizes the importance of individual creativity in the general team context. As such, a multilevel moderated mediation model (see Figure 1) is developed to explore team behavioral integrations’ influence on individual creativity from the perspective of organizational learning. The present study examines both explorative and exploitative learning as mediators between team behavioral integration and individual creativity to explore two different mechanisms of this relationship. Finally, this study examines the aforementioned relationship further by introducing individual creative self-efficacy as a boundary condition to integrate into the research model. The present research also tests whether the construct moderates the relationship between team behavioral integration and individual creativity via both subconstructs of organizational learning, explorative and exploitative learning.
The theoretical model.
Individual Creativity
Despite the fact that American psychologist Guilford first put forward the concept of creativity in 1950 and the topic has since been widely studied, there is still no universally accepted definition of the term. Guilford (1967) treated creativity as a personality trait, whereas Amabile (1996) argued that it refers to the transformation of individual ideas into tangible outcomes, such as products and services (George & Zhou, 2007). Other scholars define creativity as a process, starting from an individual’s realization of some gap or absence in their knowledge, skills or abilities. From there, the individual proactively engages in information-seeking, followed by creating original products or ideas to help solve or alleviate the existing gap (Torrance, 1963). For the purpose of this study, the present research uses this operational definition of individual creativity, focusing on the process-oriented model.
Existing studies have largely explored the antecedents of individual creativity from the situation-interaction perspective, which demonstrates that individual creativity is the product of the interaction between individual characteristics and their surroundings (Woodman et al., 1993). Individual factors such as self-efficacy (Hirst et al., 2018), cognitive style (Lomberg et al., 2017), and proactive personality (Gong et al., 2009; Yang & Chau, 2016) have been linked to influence individual creativity. Situational factors like leadership style (Qu et al., 2015), organizational culture (Shao, 2019), and team characteristics (Naranjo-Valencia et al., 2017; Yoon et al., 2016) also influence creativity. Both individual and situational factors influence employee creativity. While individual creativity has been well studied, more research is needed to better understand the mechanisms that influence individual creativity in the context of working in a team.
Team Behavioral Integration and Individual Creativity
Individual creativity as a result of various environmental factors, including social and team integration, was first introduced by Hambrick and Mason (1984) within the framework of upper echelon theory. They prescribed that behavioral integration reflects the degree of convergence of team member interactions, mainly upper-level management, and demonstrates their collective behavior. There are three dimensions of behavioral integration: quality and quantity of team information exchange, teamwork behavior, and collective decisions. Each dimension reinforces and promotes the other and explains how the team operates and works together. Given that Hambrick’s original definition is still most often cited (Ling et al., 2008; Simsek et al, 2005), it will be adapted from upper management teams to general management teams for the framework moving forward in this paper.
As one of the most important behavior processes of team members actively sharing information and resources (Yao & Sun, 2010), team behavioral integration has been shown to influence many outcome variables at several levels. In terms of organizational performance, high team behavioral integration enables enterprises to quickly grasp relevant information and respond to market changes, allowing them to make high-quality strategic decisions that improve organizational performance (Liu & Jiang, 2015; Rosenkranz & Wulf, 2019). Moreover, the dimensions of behavioral integration have a significant impact on organizational innovation. Teams with high behavioral integration generate more communication through cooperative behavior, information sharing, and joint decision-making, thus integrating knowledge more efficiently and improving organizational performance (Liu & Jiang, 2015). Despite most research on behavioral integration focusing more on top management teams, newer studies have found that team behavioral integration is positively related to innovation performance in lower-level teams, such as R&D teams (Liu & Liu, 2012). For individual outcomes, most of the research on behavioral integration focuses on its impact on personal skills, employees’ job satisfaction, and entrepreneurship (Raes et al., 2013).
Thus, prior research contends that team behavioral integration positively influences individual creativity within teams, which in turn may be related to individual performance. The cognitive processes related to creativity include problem identification, information search, and creativity generation (Cramond et al., 2010). That is, when individuals invest more time and energy in identifying problems, searching for more information, and putting forward more ideas, they are more likely to generate novel and practical ideas. Teams with high behavioral integration are characterized by open and timely communication of information among team members, habitual teamwork and joint decision-making (Hambrick, 1987). This enables team members to obtain more valuable information, knowledge, and resources. Moreover, information exchange and teamwork brought by team behavioral integration enhances employees’ confidence in completing tasks and creates a more positive working atmosphere. Another dimension of team behavioral integration, collective decision making, helps team members participate more fully in the decision-making process, therefore strengthening a greater sense of ownership for employees while helping them form a closer connection with the organization. Lastly, a team with higher behavioral integration tends to perform in more diverse ways, which then enriches and expands members thinking patterns, improving individual creativity as a result (Amabile et al., 2005). Thus, the following hypothesis is proposed:
Organizational Learning as a Mediating Factor
Scholars have contributed to the understanding of organizational learning from different disciplines and perspectives, such as economics, sociology, and human resource management. Organizational learning is a dynamic process in which organizations actively acquire information and adjust their behavior in the face of rapidly changing business environments (Argyris & Schon, 1978). Based on the classic dimension division by March (1991), organizational learning is split into explorative learning and exploitative learning. Explorative learning refers to identifying and understanding new knowledge of potential value outside the enterprise, while exploitative learning refers to the creation of new knowledge and business output from existing knowledge already within the organization (Lane & Lubatkin, 1998). Explorative learning is an organization’s propensity to become more inclined to learn new knowledge and skills, while exploitative learning is an organization’s propensity to become more inclined to use experience and improve existing knowledge. The present research focuses on these two sub-dimensions of organizational learning rather than looking at the construct as a whole.
As an antecedent variable, organizational learning has been found to enable enterprises to have stronger innovation capabilities (Zhang & Yu, 2015). Exploitative learning is positively correlated with incremental innovation of enterprises, while explorative learning can significantly promote fundamental innovation of enterprises (Cohen & Caner, 2016). Prior studies indicate that organizational learning institutionalized through multi-level initiatives promotes individual learning, which in turn further enhances organizational learning, thereby improving organizational dynamic ability. Organizational dynamic ability can also be reflected in the process of employees engaging in explorative learning, internally adjusting themselves to deal with the external environment by constantly creating new knowledge (Cepeda & Vera, 2007). As an outcome variable at the individual level of analysis, organizational learning is influenced by intellectual capital and social capital. For example, when employees have high social capital, they can come into contact with all kinds of knowledge sources, thus, employees have the capacity to improve organizational learning ability after effective absorption of information (Hsu & Fang, 2009).
Teams with high behavioral integration can better apply, promote, and acquire knowledge. Although team behavioral integration can enhance creativity, the process depends upon individual learning ability. Thus, the present study proposes that organizational learning at the individual level of analysis mediates the relationship between team behavioral integration and individual creativity. According to social learning theory, adults are inclined to learn according to their needs and interests (Bandura, 1977), so team behavioral integration fosters the appropriate atmosphere and motivation conducive for employees to learn (Illeris, 2003). Information exchange and better teamwork among members that occurs through positive team behavioral integration enables team members to interact with other members frequently and get direct and indirect experience for individual learning. Moreover, through promoting cooperation and high-quality information exchange, behavioral integration generates social mechanisms such as trust, dispels consideration of team members’ reluctance to share, and stimulates a degree of creative professional discussion and information knowledge sharing (Carmeli & Schaubroeck, 2006), which is important for explorative learning. Exploration learning enables employees to learn new knowledge and methods, which is a kind of innovation support for employees, thus promoting their creativity. Therefore, the following hypothesis is proposed:
Given that exploitative learning pertains to seeking new information through the collection of existing information or existing thinking patterns, team behavioral integration characterized by information exchange and teamwork is conducive to exploitative learning activities. Through exploitative learning, old knowledge and new knowledge can merge with each other, as team members can constantly learn from each other’s knowledge and skills, and work with each other to create new ideas, which can then improve employees’ sense of self-efficacy and stimulation of their creativity. Therefore, the following hypothesis is also proposed:
Creative Self-Efficacy as a Moderating Factor
Creative self-efficacy is defined as one’s ability to exhibit self-confidence when performing creative tasks (Bandura, 1977; Tierney & Farmer, 2002). Individuals with high levels of creative self-efficacy are more engaged in learning processes as a way to promote their own creativity and ability in completing tasks. Creative self-efficacy influences individuals’ judgments of their own KSAO’s (i.e., knowledge, skills, abilities, and other characteristics), perceived confidence, and expectations about creativity outcomes (Bandura, 1986). Several studies have confirmed the positive role of creative self-efficacy in promoting individuals’ endeavor in the creativity process (Diliello et al., 2011; Gong et al., 2009; Mittal & Dhar, 2015; Wang et al., 2014).
As previously mentioned, organizational learning is one of the frameworks in which dynamic learning processes occur within rapidly changing business environments (Argyris & Schon, 1978). Like creative self-efficacy, organizational learning therefore requires a certain level of confidence in existing knowledge, combined with the prototyping and generation of new ideas (Ford, 2000). Therefore, this study asserts that an individuals’ creative self-efficacy helps to facilitate the organizational learning process.
More specifically, explorative learning requires high levels of creative self-efficacy in learning new knowledge and skills which may be unfamiliar and challenging. Prior studies have indicated that individuals with high levels of creative self-efficacy exhibit higher levels of confidence, and this perception of confidence allows those who seek new knowledge to believe it is a sign of improvement rather than a sign of lacking knowledge (Edmondson, 2004). Moreover, it further increases individual creativity by allowing individuals to experiment with new ideas (Earley & Lituchy, 1991). Likewise, exploitative learning requires a certain level of confidence and humility about first acknowledging, then subsequently correcting the limitations of their existing KSAO’s. Individuals with high levels of creative self-efficacy enact exploitative learning because it encourages them to review existing knowledge, such that reflections from past experiences can be used toward a more positive reframing of their KSAO’s. Additionally, individuals with higher creative self-efficacy engage more in exploitative learning to merge old knowledge with new and emerging knowledge that is occurring within an organization, specifically to enhance creative performance (Tierney & Farmer, 2002). Therefore, individuals’ creative self-efficacy supports the organization learning process (explorative learning and exploitative learning). The present research proposes that creative self-efficacy strengthens the mediating role of organization learning on the relationship between team behavioral integration and individual creativity:
Study 1
Participants and Procedure
In Study 1, general management teams in large technological enterprises were selected (e.g., Dianping, Baidu), located in Beijing, Shanghai and Chengdu Province in China. Respondents were randomly selected in collaboration with the HR department staff of the surveyed enterprises. Employees were aggregated into teams based upon their respective functional operation unit. Employees were then grouped with their direct report unit leader. This process led to a total of 92 general management teams, each team comprising of one leader and an average of 3 to 10 team members.
In order to avoid common method bias, data was gathered from different sources and at different times (Podsakoff et al., 2003). Two separate questionnaires were designed for leaders and subordinates. Each questionnaire was assigned an identification number so the responses of the subordinates could be matched with the evaluations of the immediate leaders. A cover letter accompanying the questionnaire indicated that the survey was being conducted solely for academic research purposes and the participants were assured of the confidentiality of their responses. Paired questionnaires were administered at two time points. Employees were asked to complete survey items that pertain to team behavioral integration, their own level of organizational learning, individual creative self-efficacy, and demographic characteristics (e.g., age) at Time 1 (May, 2019). Team leaders rated their subordinates’ creativity at Time 2 (June, 2019).
A total of 92 teams were invited to participate in the study, including 92 leaders and 387 team members; 88 leader and 367 subordinate questionnaires were returned with a response rate of 95.7% and 94.8%, respectively. After omitting incomplete and unmatched questionnaires, a total of 86 leader and 356 subordinate questionnaires were obtained with an effective completion rate of 93.5% and 92.0%, respectively.
Men accounted for 47.19% and women 52.81%. Employees aged 26 to 35 years old accounted for the largest age proportion (55.61%). 87.08% of team members comprised of undergraduate and postgraduate degree holders. There were more male team leaders than female, with male leaders accounting for 73.17% of total leadership. Of this leadership population, 48.78% held advanced or graduate degrees. The average tenure of team leaders was 11 years within each respective organization.
Measures
Mature scales were adapted in the questionnaire design and used 6-level Likert scale ratings (from 1 = strongly disagree to 6 = strongly agree) to avoid a neutral response (Chen et al., 1995). Upon suggestion from subject matter experts’ evaluation and a team of eight graduate students, items with poor face validity were revised due to unclear and ambiguous language translations that occurred during the adaptation of the survey from Chinese to English. Ten teams of graduate MBA students at Shanghai International Studies University were selected to participate in an initial pilot survey to further assess the translated items. Subsequently, the items were revised again according to the initial measurement results which led to the final version of the formal questionnaire.
Team behavioral integration
The team behavioral integration scale developed by Li and Hambrick (2005) was used to measure team behavioral integration. Four items measure three dimensions of team behavioral integration: information exchange, cooperation, and collective decision-making. A sample item is “All team members have a voice in team decisions.” Cronbach’s α for the scale is .88. Team behavioral integration is a variable observed at the team level; thus, the collected data at the individual level was pooled at the team level for analysis. We adopted the mean method approach to integrate the collected individual data of behavioral integration into team data. The Rwg indicator was used to evaluate the intra-group consistency of this variable, and ICC (1) and ICC (2) indicators were used to evaluate inter-group heterogeneity (James et al., 1984). According to the results, ICC (1) = .38 (>.05), ICC (2) = .73 (>.50), and mean Rwg(j) = .80 (≥.70), data met the aggregation conditions, and highlighted that the data can be aggregated to the team level.
Organizational learning
Organizational learning was measured with Atuahene-Gima and Murray’s (2007) 11-item scale. Six items measure explorative learning and five items measure exploitative learning. Sample items includes: “I often enter new areas by learning new knowledge” (explorative learning); “I integrate existing knowledge to improve existing products” (exploitative learning). Cronbach’s α for explorative learning and exploitative learning subscale are .89 and .92, respectively.
Creative self-efficacy
Creative self-efficacy was measured using the 3-item scale developed by Tierney and Farmer (2002) to measure individual creative self-efficacy. A sample item is “I feel that I am good at generating novel ideas.” Cronbach’s α is .90.
Individual Creativity
Farmer et al.’s (2003) 4-item scale to measure individual creativity. A sample item is “This team member would seek new methods to solve problems.” Team leaders were asked to rate team members’ creativity at time 2. Cronbach’s α is .91.
Control Variables
Demographic information was collected (i.e., gender, age, education) and team size as control variables because previous research indicated that employee demographic characteristics may be associated with creativity (e.g., Zhou & Shalley, 2008), and team size may be associated with team process (e.g., Li & Hambrick, 2005). Some variables were coded as a dummy variable: gender (1 = male, 2 = female), age (1 = less than or equal to 25, 2 = 26–35, 3 = 36–45, 4 = greater than or equal to 45), education (1 = junior college diploma or below, 2 = bachelor, 3 = master, 4 = doctor or above).
Results
Given that data were collected for team behavioral integration, explorative learning, exploitation learning, and creative self-efficacy at the same time point in Study 1, an initial analysis inspecting possible common method bias problems was performed. A Harman’s one-factor analysis was conducted following the suggested procedure by Andersson and Bateman (1997) by including all indicators of the four variables and fixing to one without rotation. The factor generated explained only 30.74% of the total variance, which is lower than the 50% threshold, suggesting that common method variance is not a major concern.
A series of confirmatory factor analyses (CFAs) were conducted at the individual level to explore the distinctiveness of the focus five variables. As shown in Table 1, the hypothesized five-factor model (X2/df = 4.13, SRMR = .05, RMSEA = .05, CFI = .90, TLI = .89) fit the data better than alternative models, providing support for the distinctiveness of the five constructs in the current study.
Comparison of Measurement Models.
Note. N (individuals) = 356 for study 1; N (individuals) = 138 for study 2; TBI = team behavioral integration; EL1 = Exploitative learning; EL2 = exploitative learning; CSE = creative self-efficacy; IC = individual creativity.
Represents two factors merged into one.
Descriptive Statistics and Correlations
Table 2 shows the means, standard deviations, inter-correlations, and reliability coefficients of the variables and highlights the positive correlation between these five constructs. These bivariate results provide preliminary support for the hypothesized relations.
Descriptive Statistics and Correlation among Variables.
Note. N (individuals) = 356 for Study 1, N (individuals) = 138 for Study 2; TBI = Team Behavioral Integration, EL1= Exploitative Learning, EL2 = Exploitative Learning, CSE = Creative Self-efficacy, IC = Individual Creativity.
p < .05. **p < .01. ***p < .001.
Hypotheses Testing
Given that the data used in this study is nested (employees nested within teams), MPlus 7 (Muthén & Muthén, 2012), an integrated path-analytic approach, was used to test multilevel mediation and moderated mediation hypotheses for this research (Preacher et al., 2010). A null model without any predictors to assess the significance of between-group variance in organizational learning scores was calculated. Results of the null model showed the ICC(1) value were .104 and .110, suggesting that approximately 10.4% of the variance in individual level ratings of explorative learning and 11.0% variance of exploitative learning came from group membership. Likewise, another null model without predictors to assess the significance of between-group variance in individual creativity scores was tested. The ICC(1) value of .231 confirmed that 23.1% of the variance in individual level ratings of individual creativity came from group membership. The ICC(1) value was above the acceptable median value of .059, and thus justified multilevel analyses (Cohen, 1988). The hypothetical model was then tested. The model exhibited acceptable fit to the data, X2/df = 5.04, SRMR = .06, RMSEA = .07, CFI = .88, TLI = .87; a depiction of the results of the multilevel path analysis is shown in Figure 2. Age, gender, education, work year, and team size were all included as controls with fixed effects on organizational learning and individual creativity. Supporting Hypothesis 1, team behavioral integration was positively related to individual creativity (γ = .60, p < .001).
Path analysis results for study 1.
Hypotheses 2a and 2b predicted that explorative learning and exploitative learning would mediate the relationship between team behavioral integration and individual creativity, respectively. A bootstrap procedure was used to assess the hypothesized indirect relationship (Preacher et al., 2010). The indirect effect is significant when the 95% confidence interval of Parameter-based Bootstrap does not contain zero. With 2,000 Monte Carlo replications, the results show that there is positive indirect relationship between team behavioral integration and individual creativity via explorative learning (indirect effect = .104, 95% CI excluded 0: [.067, .841]). Therefore, H2a was supported. Similarly, there is positive indirect relationship between team behavioral integration and individual creativity via exploitative learning (indirect effect = .075, 95% CI excluded 0: [.023, .762]); thus, H2b was also supported.
Hypotheses 3a and 3b predicted that individual creative self-efficacy moderates the indirect relationship between team behavioral integration and individual creativity via organization learning. To assess these moderated mediations (Muller et al., 2005; Preacher et al., 2010), four conditions were examined: (1) the significant effect of team behavioral integration on individual creativity; (2) the significant effect of organizational learning (i.e., explorative learning and exploitative learning) on individual creativity; (3) the significant interaction between organizational learning and individual creative self-efficacy in predicting individual creativity; and (4) the different conditional indirect effects of team behavioral integration on individual creativity via organizational learning, across low and high levels of individual creative self-efficacy (the essence of moderated mediation). Moderated mediation is demonstrated when this conditional indirect effect differs in strength across low and high levels of moderator.
Condition 1 was supported by the results for Hypothesis 1 (γ = .60, p < .001), while Condition 2 was satisfied by the results showed in Figure 2 (γ = .26, p < .001 for explorative learning; γ = .22, p < .001 for exploitative learning). Figure 2 also demonstrates that the interaction of explorative learning with creative self-efficacy was significant in predicting individual creativity (γ = .22, p < .001), while the interaction of exploitative learning with creative self-efficacy was significant in predicting individual creativity (γ = .13, p < .01). Figure 3 and 4 plots these two interactions, respectively, thus supporting Condition 3.
Explorative learning and creative self-efficacy’ s interaction effect on individual creativity for study 1.
Exploitative learning and creative self-efficacy’ s interaction effect on individual creativity for study 1.
The analysis proceeded with a further examination of Condition 4 following Preacher et al.’s (2010) bootstrap procedure. High and low levels of individual creative self-efficacy were operationalized as one standard deviation above and below the mean score of the variable. The results showed that the conditional indirect effects of team behavioral integration on individual creativity via explorative learning was significant (indirect effect = .231, 95% CI excluded 0: [.140, .433]) for individuals with high levels of creative self-efficacy versus nonsignificant (indirect effect = .054, 95% CI included 0: [−.010, .102]) for individuals with low level of creative self-efficacy. There was also a significant difference in the estimates of these two mediation effects (Δ = .177, 95% CI excluded 0: [.109, .481]) Likewise, the conditional indirect effects of team behavioral integration on individual creativity via exploitative learning was significant (indirect effect t = .203, 95% CI excluded 0: [.151, .323]) in high individual creative self-efficacy but nonsignificant (indirect effect = .031, 95% CI excluded 0: [−.040, .083]) in low individual creative self-efficacy. And there is a significant difference in the estimates of these two mediation effects (Δ = .172, 95% CI excluded 0: [.017, .326]). Thus, H3a and H3b were supported.
Study 2
A second study was conducted in order to glean additional insights into (1) the high correlations between explorative learning and exploitative learning, and (2) the causal relationship between team behavioral integration and organizational learning. Team members from the previously surveyed teams were first selected to conduct deep interviews to discern how they understood the different learning patterns. Interviews revealed that employees need to apply both learning patterns in their daily life due to the nature of their jobs (i.e., working in high-tech enterprises). Therefore, in the follow-up study, general management teams from other industries were examined, including manufacturing, service, and other companies.
Additionally, the causal relationship between team behavior integration and organizational learning was addressed. Study 1 collected data from both variables at the same timepoint, leading to potential common method variance bias. As such, the second study was distributed over three time points.
Participants and Procedure
In Study 2, general management teams were chosen in several enterprises involving machinery, electronic communication and service, mainly located in the Shanghai and Zhejiang Provinces in China. Similar to Study 1, in order to avoid common method bias (Podsakoff et al., 2003), data was gathered from different sources and at different times in Study 2. Two separate questionnaires were designed for leaders and subordinates and distributed at three time points. At Time 1 (February, 2020), employees were asked to rate their perception of team behavioral integration and their demographic information. At Time 2 (May, 2020), employees filled out survey items including their own level of organizational learning and creative self-efficacy. At Time 3 (July, 2020), team leaders evaluated their subordinates’ individual creativity.
Questionnaires were distributed to a total of 55 teams, including 55 leaders and 231 team members. After omitting unmatched questionnaires, 39 leader and 138 subordinate questionnaires were obtained with an effective completion rate of 70.9% and 59.7%, respectively. The sample was 53.4% male and 40.39% female; 52.59% members were between the age of 26 and 40; 86.52% of team members were undergraduate and postgraduate degree holders. The majority of team leaders were male (93.23%), and all of team leaders had higher education (51.62% held advanced or graduate degrees), with an average of 13 years of work experience.
Measures
The same mature scales used in Study 1 were also used in Study 2. Cronbach’s α for team behavioral integration was .89. The values of ICC (1), ICC (2) and mean Rwg(j) were .36, .71, and .91, respectively, indicating that team behavioral integration data can be aggregated to the team level. Cronbach’s α for explorative learning and exploitative learning were .88 and .89, respectively. Cronbach’s α for creative self-efficacy is .91; Cronbach’s α for individual creativity is .91. Demographic information (i.e., gender, age, education) and team size were collected as control variables.
Results
A series of confirmatory factor analyses (CFAs) were conducted and the results are shown in Table 1. Likewise, the hypothesized five-factor model (X2/df = 3.26, SRMR = .05, RMSEA = .05, CFI = .91, TLI = .90) fit the data better than alternative models, providing support for the distinctiveness of the five constructs in the current study. Table 2 presents the means, standard deviations, inter-correlations, and reliability coefficients of the variables based on the sample of Study 2.
To test Study 2 hypotheses, a null model without any predictors was calculated to assess the significance of between-group variance in organizational learning and individual creativity scores. Results of the null model showed that the ICC(1) value of explorative learning, exploitative learning and individual creativity were .100, .108, and .212, respectively. The ICC(1) value confirmed there was sufficient inter-group variance and justified multilevel analyses. We then tested the hypothetical model together with control variables. The model exhibited acceptable fit to the data (X2/df = 4.34, SRMR = .06, RMSEA = .06, CFI = .90, TLI = .88), and Figure 5 illustrates the results of multilevel analysis.
Path analysis results for study 2.
Supporting H1, team behavioral integration was positively related to individual creativity (γ = .32, p < .001). With 2,000 Monte Carlo replications, the results show that explorative learning moderates the relationship between team behavioral integration and individual creativity (indirect effect = .144, 95% CI excluded 0: [.102, .238]). Likewise, exploitative learning moderates the relationship between team behavioral integration and individual creativity (indirect effect = .097, 95% CI excluded 0: [.011, .143]). H2a and H2b were both supported.
H3a and 3b were assessed by examining the aforementioned four conditions (Muller et al., 2005; Preacher et al., 2010). Condition 1 was supported by the results for Hypothesis 1 (γ = .32, p < .001), while Condition 2 was satisfied by the results showed in Figure 5 (γ = .37, p < .001 for explorative learning; γ = .27, p < .001 for exploitative learning). Figure 5 also demonstrates that the interaction of explorative learning with creative self-efficacy was significant in predicting individual creativity (γ = .32, p < .001), while the interaction of exploitative learning with creative self-efficacy was significant in predicting individual creativity (γ = .24, p < .001). Figures 6 and 7 plot these two interactions, respectively, thus supporting Condition 3. As for Condition 4, the bootstrapped results showed that the conditional indirect effects of team behavioral integration on individual creativity via explorative learning was significant (indirect effect = .320, 95% CI excluded 0: [.114, .335]) for individuals with high levels of creative self-efficacy versus nonsignificant (indirect effect = .061, 95% CI included 0: [−.054, .114]) for individuals with low level of creative self-efficacy. There was a significant difference in the estimates of these two mediation effects (Δ = .259, 95% CI excluded 0: [.200, .475]). Likewise, the conditional indirect effects of team behavioral integration on individual creativity via exploitative learning was stronger and more significant (indirect effect = .211, 95% CI excluded 0: [.105, .539]) in high individual creative self-efficacy but weaker (indirect effect = .095, 95% CI excluded 0: [.078, .136]) in low individual creative self-efficacy. There was a significant difference in the estimates of these two mediation effects (Δ = .116, 95% CI excluded 0: [.023, .198]). Thus, Hypotheses 3a and 3b were supported.
Explorative learning and creative self-efficacy’ s interaction effect on individual creativity for study 2.
Exploitative learning and creative self-efficacy’ s interaction effect on individual creativity for study 2.
Discussion
In the present research, a multilevel theoretical model was constructed to explore team behavioral integration’s influence on individual creativity in general management teams, examining the mediating role of organizational learning’s sub-dimensions, and the moderating role of individual creative self-efficacy. To test the hypothesized model, two studies were conducted and data were collected from participants in different industries. The findings from Study 2 complimented the preliminary findings in Study 1 with an expanded sample and three data collection points, demonstrating that the correlation between the two learning patterns varied by industry; for example, the correlation between the two learning patterns in Study 2 (r = .69) was found to be significantly lower than that in Study 1 (r = .86). Both study results confirmed that team behavioral integration is positively related to individual creativity, and explorative learning and exploitative learning mediate this relationship. This suggests that employees’ abilities to seek and develop existing knowledge, while considering team members’ ideas and influence, is vital to their creative processes. Results also indicate that explorative learning has a stronger mediating effect than exploitative learning. This may be attributed to the fact that teams sampled required more continuous fluidity of new knowledge rather than the development of existing knowledge, such as the acquisition of new technical skills. In addition, it confirmed the purpose of clarifying explorative and exploitative learning as two separate mechanisms and predicting the strength difference of the two learning patterns’ mediating effect. Results also indicate that individual creative self-efficacy strengthens the indirect relationship between team behavioral integration and individual creativity, suggesting that self-regulatory behaviors are paramount for creative processes to occur. Employees are thus encouraged to develop an inner capacity to filter and engage with diverse forms of information that are introduced throughout the creativity process.
Theoretical Implications
In the era of the knowledge economy, more organizations recognize the importance of teams as the primary work unit. These teams not only focus on top management teams, but also on general management teams such as R&D and knowledge management teams (Mesmer-Magnus & DeChurch, 2009; Qu & Liu, 2017; Tao, 2011; Xu, 2013). Although a wealth of research has explored the importance of learning behaviors in organizations, there is limited information about which team level variables are positively related to individual learning processes. This study extends the understanding of behavioral integration by empirically testing its impact on individual creativity. These studies found that behavioral integration is important to general management teams in large Chinese technology companies, extending the empirical utility of upper echelon theory and social learning theory in cross-functional and cross-cultural team contexts in Asian samples. This study also incorporates organizational learning into the framework to discuss how behavioral integration is related to individual creativity through the combination of two learning patterns.
Additionally, the study extends the understanding of how individual self-efficacy and individual learning interact to improve employees’ creativity. Current studies on the effect of this interaction are developing, but are still limited (Bommer et al., 2003). Within a team setting, learning often occurs when individuals obtain and develop knowledge and insights from team members (Boone & Kendricks, 2009). Research on individual differences, such as variability in creative self-efficacy, is a new perspective to explore how to foster and promote individual learning behaviors (Edmondson, 2004). Based on social learning theory (Bandura, 1977), this study explains that the interaction between individual traits and situations helps to improve creativity. First, in response to the call of scholars for the creative interaction perspective, this study focuses on the important situation created by work teams for individual creativity promotion. Therefore, the construct of team behavioral integration was applied, which has been primarily studied in top management teams, as well as in general management teams. In general management teams, team behavioral integration and its three dimensions of behavior interaction create an atmosphere conducive to improving its members’ ability to learn. Team behavioral integration provides a source for two different behavior acquisition methods, so it plays an active role in exploratory learning and exploitative learning. The view of self-regulation in social learning theory explains why high creative self-efficacy has the moderated mediation effect. Creative self-efficacy, as an individual trait related to creativity, proves that the interaction between individual trait and situation positively affects creativity.
Practical Implications
These studies reveal several insights into how team members can facilitate individual creativity through individual learning and individual characteristics of creative self-efficacy. Organizations can apply findings through the deliberate facilitation of team behavioral integration, such as team-building exercises, team development, and experiential programs, in both in person and online formats. Drawing on the study findings, the present research contends that the formation of high-efficiency teams requires a higher degree of behavioral integration. First, it is advisable to enhance cooperative behaviors within a team. For example, organizations can divide a project assignment based on employees’ strengths, using goal setting tools to strengthen communication among team members (Lubatkin et al., 2006). Second, facilitating information exchange among team members helps foster team behavioral integration. In high functioning teams, team leaders can regularly let members share their recent work and exchange knowledge and experiences to help in this regard. Finally, leaders can allow team members to make decisions together. When making decisions, team leaders should encourage team members to express their opinions and actively put forward their own suggestions, thereby allowing team members to become active participators (Carmeli & Schaubroeck, 2006).
Additionally, explorative learning and exploitative learning go hand in hand. On the one hand, team members should be encouraged to learn new information and have opportunities to acquire more knowledge, such as setting up a learning group to share knowledge and skills regularly, or inviting other companies’ exemplar team members to share their experiences (Dixon, 2017). Organizations need to encourage team members to broaden their ways of acquiring knowledge and conducting research. For example, team members may regularly organize and report the technological innovation information of other companies in their markets (Fernández-Mesa & Alegre, 2015). While doing so, team members should be encouraged to also strengthen the integration of existing knowledge within their organization, such as setting up cross-team-building events to learn about other teams’ techniques and methods (Mostafa & Klepper, 2017). Writing and sharing regular internal summaries can be an excellent way for employees to integrate existing knowledge and technology. Both team leaders and team members are invited to adopt organizational learning, and reasonably allocate time to facilitate both explorative and exploitative learning.
Furthermore, managers can improve employees’ creative self-efficacy by providing a work environment that encourages team members to innovate with new behaviors (e.g., innovation awards and internal contests) and recognize creative potential to help instill confidence in employees to fulfill new goals (Gong et al., 2009; Tierney & Farmer, 2002). Organizations can build individual creative self-efficacy and individual learning skills by promoting peer-to-peer learning and coaching (Dixon, 2017). Finally, in the teamwork environment, team leaders also play an important role in this process. Team leaders can model self-regulatory creative processes, and verbally walk through their thought processes. This will persuade employees to engage in creative behaviors without fear of reappraisal (Gong et al., 2009). Support from team leaders can boost employees’ creative self-efficacy, and thereby increase individual creativity by enhancing the individual learning behaviors.
Research Limitations and Future Prospects
As with all research, these studies have several limitations that could be addressed by future research. First, the correlation between the sub-dimensions of learning, explorative learning and exploitative learning, was found to be high in Study 1. This high correlation is not rare (e.g., Qu & Liu, 2017), but it is believed that this particularly strong correlation was an artifact of the sample. Study 2 aimed to address this issue, expanding beyond participants working in technology to those working in an array of other industries, and found that the correlation between the learning dimensions was not as strong. Future research should continue to explore the differences between explorative and exploitative learning in other industries and illustrate the best cut-off of 95% for the goodness of fit index CFI.
While the studies assert that behavioral integration is related to explorative and exploitative learning, the present research cannot determine causality. Three timepoints were used in Study 2 to demonstrate that behavioral integration can lead to explorative and exploitative learning, but that does not negate the possibility of the learning dimensions further influencing behavioral integration. Given that the variables may occur on a circumplex (i.e., subconstructs are not orthogonal), future research could assess whether an upward spiral of learning and behavioral integration exists, where behavioral integration leads to increased explorative and exploitative learning, which then leads to even greater behavioral integration (Epskamp et al., 2012).
Another future research direction could focus on team integration concepts outside of behavioral integration, such as affective integration (i.e., how teammates perceive the quality of their interpersonal relationships within the team) and cognitive integration (i.e., the amount teammates comprehend each other’s interpretive frameworks; Cronin et al., 2011). Understanding how affective and cognitive integration specifically impact explorative and exploitative learning and individual creativity, particularly in virtual and online formats, can help expand upon the benefits of different team-focused integration concepts. Moreover, consideration of cross-cultural teams may help the constructs of team behavioral integration and creative self-efficacy generalize across a global population (Donaldson et al., 2020; Villalobos et al., 2020).
This study assesses the influential mechanism of team behavioral integration on individual creativity from the perspective of organizational learning. However, organizational learning is just one of the mechanisms that can bridge team behavioral integration and individual creativity. Future research may seek to start from other perspectives, such as motivational behavior theories (e.g., psychological safety) or use an emotion perspective (e.g., positive emotions; Hu et al., 2018; Yang & Hung, 2015). Similarly, there are other self-regulatory behaviors besides individual self-efficacy that influence how individuals interact with their surroundings. Future research may consider other individual characteristics, such as proactivity (Carmeli & Shteigman, 2010).
Conclusion
As organizations increasingly rely on teams and team-based decision making, it is important that they pay more attention to the behavioral interactions among team members to better understand the antecedents and mechanisms of individual creativity. Based on social learning theory, this study confirms the multilevel mechanisms between team behavioral integration and individual creativity from an organizational learning perspective and demonstrates the boundary effect of individual creative self-efficacy. These findings help paint a more comprehensive picture of the effect of team behaviors on individual creativity.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This project was supported by the National Natural Science Foundation of China (NSFC): This project was supported by the National Natural Science Foundation of China (NSFC). Project number: 71702105; 71701131.
