Interdisciplinarity and Team Innovation: The Role of Team Experiential and Relational Resources

Abstract

Interdisciplinary teams composed of members with different expertise possess a variety of perspectives, which increases their potential for innovation. In reality, team members often fail to integrate their expertise, resulting in the team not reaching its innovative potential. It is argued the unsharedness of expertise within interdisciplinary teams has an inverted-U relationship with innovation. To explore the conditions under which the unsharedness of expertise enhances or impairs innovation, the resource-based view of organizational productivity is applied to teams. It is argued the amount and configuration of team relational and experiential resources facilitate teams’ ability to integrate members’ expertise for innovation.

Keywords

team expertise team cognition interdisciplinary teams team innovation team creativity

Given the rate of new-knowledge creation and the concomitant need for specialization, innovation has increasingly become a function of interdisciplinary teams (Ahmadpoor & Jones, 2019; Fiore, 2008; Salazar et al., 2012). Interdisciplinary teams are composed of experts with unique expertise that are assembled to generate new understandings of existing problems and to explore new opportunities (Huckman et al., 2009; Leahey et al., 2017; Lee et al., 2015). In a variety of areas, from new product development, engineering, and scientific research, interdisciplinary teams are recognized as catalysts of learning and innovation (Edmondson, 2002, 2012; Edmondson & Nembhard, 2009). Despite their creative potential, research suggests interdisciplinary teams may struggle to integrate their expertise into a unified whole, and thus not reach their innovative potential (Mesmer-Magnus & DeChurch, 2009; Salazar et al., 2012; Xiao et al., 2016). As such, research on team innovation and the question of why some teams are more innovative than others remains a central focus of teams research (Hoever et al., 2018; Keyton & Heylen, 2017; van Knippenberg, 2017).

The present study advances research on the innovative potential of interdisciplinary teams by exploring how the degree of expertise unsharedness among team members relates to team innovation. Applying the resource-based view (RBV) of organizational productivity to teams, it is argued that both team experiential (accumulated team member expertise) and relational (extent to which members of the team have prior experience working together; team familiarity) resources determines the team’s “instrumental expertise”—that expertise which can be leveraged for innovation. As such, a team’s instrumental expertise is a function of the unsharedness of expertise among team members as well as the composition of team member resources (the additive elements of team experience and familiarity) and the nature of its compilation (the extent to which experience and familiarity is concentrated or shared across the team; DeChurch & Mesmer-Magnus, 2010; Kozlowski & Klein, 2000).

Interdisciplinary Teams and Innovation

Interdisciplinary teams with varied expertise and training have the potential to recombine their knowledge in novel ways to create truly innovative outcomes (Taylor & Greve, 2006; West, 2002; Yong et al., 2014). Teams research suggests (a) teams composed of members with different knowledge and expertise have greater innovative potential (Mello & Rentsch, 2015; van Dijk et al., 2012; van Knippenberg, 2017), and (b) realizing this potential depends upon teams’ ability to share and integrate distinct member expertise in pursuit of team goals (Baumann & Bonner, 2013; Beck et al., 2017; Lehmann-Willenbrock et al., 2017). Research also suggests that teams are not naturally adept at sharing and integrating their knowledge resources (Mesmer-Magnus & DeChurch, 2009; Stasser & Titus, 1985). The implications of these findings are that innovation in interdisciplinary teams is not only a function of the differences in their combined knowledge, but also their success in integrating it at the team-level (Rico et al., 2008; Xiao et al., 2016).

Team innovativeness—the ability to generate novel and useful solutions, inventions, or ideas—is a recombinatorial process (Benoliel & Somech, 2015; van Knippenberg, 2017), so differences in team members’ expertise present an opportunity for novel solutions (Fleming, 2001; West, 2002). If team members share similar expertise, then the knowledge among members is largely redundant and the team may lack the requisite variety needed for effective recombination (Fleming, 2001). Conversely, if the team members have little expertise in common, the members will struggle to capitalize on their potential because team members will not possess the shared knowledge necessary to absorb and make use of one another’s expertise (Cohen & Levinthal, 1990; Mathieu et al., 2000).

Researchers have tended to describe the differences in expertise among team members in terms of the diversity of expertise held within the team, operationalizing expertise differences in terms of functional diversity, or the distribution of members across a range of different knowledge domains (cf. Bell et al., 2011; Harrison & Klein, 2007). In this tradition, teams are considered diverse when, as a whole, the team possesses expertise that covers a variety of different domains (Harrison & Klein, 2007). However, this conceptualization of expertise as “variety” is incomplete in two important ways. First, whereas functional diversity is informative in some contexts (e.g., describing the distribution of top management team [TMT] members across functions such as marketing, human resources, and operations), it does not tell the whole story. For example, a team of chemists may, on the surface, appear to be homogeneous and have relatively little diversity of expertise. However, these chemists may specialize in one or more micro areas of chemistry, such as resins, organic compounds, or gasses. Thus, conceptualizing expertise as variety does not capture the breadth of relevant expertise each member brings to the team (Bunderson & Sutcliffe, 2002).

Second, conceptualizing expertise as variety overlooks the extent to which members’ expertise is unique rather than redundant with one another, something research on interdisciplinary teams suggests is central to these teams’ potential (Fiore, 2008; Salazar et al., 2012). For example, an interdisciplinary team may have members who have a wide breadth of expertise, and as a whole, the team possesses a diversity of expertise. However, in the aggregate, individual members’ knowledge may be largely redundant with others on the team (i.e., high variety but high overlap). Clearly, this would be less preferable from an innovation standpoint than would a team whose members’ expertise spans a variety of knowledge domains, but their knowledge is complementary rather than redundant (i.e., high variety and low overlap). Importantly, whereas teams with high expertise overlap may struggle to come up with new ideas, teams with little overlap may have difficulty utilizing their disjointed expertise. Although there is a relationship between the variety of expertise held by team members and the degree of overlap in that expertise, the two concepts clearly provide unique though complementary information. As such, the degree to which members differ in their areas of expertise, yet have enough shared understanding to know how to utilize other members’ expertise, determines the team’s ability to leverage that expertise for innovation.

To explore how and when the unsharedness in expertise among team members enhances team innovativeness, this study brings research from the RBV and dynamic capabilities to the team level (cf. Gardner et al., 2012). The RBV has traditionally been used to predict organizational-level outcomes and suggests that firm resources and capabilities, as well as the ability to deploy them, are key to predicting firm-level performance (Wernerfelt, 1984). The RBV literature is relevant to the study of interdisciplinary teams in that team resources and capabilities are predictive of team-level performance (Bell et al., 2011). A stream of research within the RBV literature on “dynamic capabilities” is particularly useful to the study of innovation within interdisciplinary teams. Dynamic capabilities in a team context refer to the team’s ability to create new knowledge from existing knowledge and include routines that facilitate the integration of individuals’ distinct expertise to improve performance (Grant, 1996; Kogut & Zander, 1992). Research in this area has suggested that a team’s capability for integrating member knowledge may be the result of teams’ “experiential” and “relational” resources, and this capability may explain why some teams are more innovative than others (Gardner et al., 2012; Xie et al., 2014). Leveraging these key insights, this study puts forth a theoretical model that explores how the amount and configuration of experiential and relational resources are associated with interdisciplinary teams’ ability to leverage the differences in members’ expertise to enhance innovation. Figure 1 depicts the theoretical model.

Figure 1.

Theoretical model.

Team Expertise and Innovation

Research suggests that combining similar knowledge domains reduces uncertainty in outcomes but can lead to innovations that are more incremental in nature and thus less innovative (Fleming, 2001). Teams whose members hold similar expertise operate from a shared knowledge base and can more easily appreciate and integrate each other’s contributions (Cohen & Levinthal, 1990). Members of such teams will have similar understandings of their technologies, shared languages and jargon, and common skills and knowledge, all of which facilitate communication and learning (Cronin & Weingart, 2007). Also, they are aware of the challenges they may face and approach these problems from similar perspectives. Similarities in expertise ensure the problem space is well understood and team members know which ideas are likely to fail and thus avoid them (Fleming, 2001). However, while the sharing and integration of similar expertise domains are easier, the upside potential is limited (Fleming & Sorenson, 2004). When team member expertise is largely redundant, any possible new-knowledge recombinations are quickly exhausted, ultimately resulting in only modest improvements to established solutions. When team members hold similar expertise, the ease of learning and knowledge sharing yields more immediate returns and makes experimentation and the pursuit of new directions less attractive (Levinthal & March, 1993).

By contrast, in teams where members possess expertise that is distinct from one another, the number of combinatorial possibilities increases, minimizing the problems of redundancy and increasing the team’s innovative potential (cf. Fleming, 2001). Having distinct areas of expertise encourages a broader search for solutions, enables the cross-fertilization of ideas, stimulates the exploration of task-relevant information (van Knippenberg et al., 2004), and increases the chances of solving particularly challenging problems.

However, research on team diversity suggests that teams wherein members’ expertise is considerably different face unique challenges. In particular, a lack of a common knowledge base and shared understanding makes it difficult to share and integrate different areas of expertise (Carlile, 2004). Teams composed of members with very different expertise approach projects from different thought worlds, leading them to frame and interpret contributions in dissimilar ways (Dougherty, 1992). Furthermore, when the differences in expertise among members are high, the number of combinatorial possibilities can be overwhelming, potentially leading to unsystematic recombination attempts and satisficing (Fleming, 2001; Fleming & Sorenson, 2004).

As such, this study argues that teams composed of members with very similar bases of expertise lack the variety of ideas needed for innovation. Conversely, teams composed of members with very different bases of expertise lack the shared understanding needed for mutual learning and thus are not able to capitalize on their potential. Rather, innovative potential is maximized when the unsharedness of team member expertise is moderate. Thus,

H1: There is an inverted U-shaped relationship between team expertise unsharedness and team innovation.

In the next section, this study describes how a team’s ability to leverage their unique expertise is a function of both the amount and configuration of their relational and experiential resources.

The Amount and Configuration of Relational and Experiential Resources

Several well-established lines of teams research offer insights as to how teams can effectively share and integrate member expertise: (a) transactive memory systems (e.g., Lewis, 2003; Z. X. Zhang et al., 2007), (b) shared mental models (e.g., DeChurch & Mesmer-Magnus, 2010; Mathieu et al., 2000; Mohammed et al., 2000), and (c) information sharing (Mesmer-Magnus et al., 2011; Rico et al., 2008; Stasser & Titus, 1985). Common to these perspectives is the notion that outcomes are better when team members communicate efficiently and have both the motivation and ability to integrate knowledge across members (De Dreu, 2007; Mathieu et al., 2000). Building on this prior work, the present study draws on research on combinative capabilities and knowledge integration in teams (Gardner et al., 2012; Kogut & Zander, 1992), and argues the amount and configuration of team relational and experiential resources (i.e., the team’s compositional and compilational expertise; Kozlowski & Klein, 2000) are important for understanding knowledge sharing and integration in teams.

Amount of Team Relational and Experiential Resources

Team relational resources refer to knowledge that is acquired by working together in the past whereby members are aware of the content and applicability of one another’s expertise (Srivastava & Gnyawali, 2011). Team cognition research might dub relational resources as transactive memory or compilational expertise, which is developed as team members have more experience working together (Kozlowski & Klein, 2000; Lewis, 2003), and refers to team members’ knowledge of who knows what in the team as well as how to capitalize on one another’s expertise. Relational resources enhance teams’ ability to integrate the different expertise held by members because they enhance the (a) efficiency of communication among members, (b) motivation for members to share information, and (c) collaborative dynamics within the team.

First, relational resources promote the development of a common language and vocabulary, allowing team members to better understand each other (Cramton, 2001). Shared terminology facilitates efficient communications among members and the effective coordination of team tasks. A common language leads to “common ground” where team members gain an awareness of their knowledge in the context of what others know. Knowing what others know allows team members to coordinate communication flows, interject at the right times, and minimize superfluous interactions (Espinosa et al., 2007). Second, greater relational resources encourage members to engage in the exchange process because they anticipate being understood (Bunderson & Sutcliffe, 2002). Prior collaborations also promote cohesion and increase motivation to share expertise, ultimately facilitating the integration of different perspectives (Reagans et al., 2005; Tzabbar & Vestal, 2015). Relational resources further improve the integration of member expertise because they encourage members to exert more effort by reframing contributions and ensuring what was shared was understood (Uzzi & Lancaster, 2003). Third, prior team experience can also provide the social support needed to experiment and take risks in the recombination of distant perspectives, ultimately improving team performance (Espinosa et al., 2007; Reagans et al., 2005). As team members gain experience working together, they build trust (Uzzi & Lancaster, 2003; Uzzi, 1997). Team environments characterized by trust foster the freedom to seek feedback and help, discuss mistakes, challenge prevailing wisdom, and experiment (Edmondson, 1999). In addition, prior collaborations among team members can reduce the likelihood of faultlines developing along similar backgrounds and training (Polzer et al., 2006), and lessen the negative effects of surface-level diversity on team functioning (e.g., self-categorization; Harrison et al., 2002).

In sum, relational resources enable the development of a shared language and common ground, increase the motivation of team members to share and exchange expertise, and improve the collaborative dynamics on the team, all of which offset the challenges associated with integrating disparate areas of expertise and improve teams’ ability to leverage their expertise to enhance innovation. Thus,

H2: Team relational resources positively moderate the relationship between team expertise unsharedness and team innovation such that expertise unsharedness more strongly predicts innovation as team relational resources increase.

Team experiential resources—the accumulation of team member expertise (the shared cognition literature might dub this compositional expertise; DeChurch & Mesmer-Magnus, 2010; Kozlowski & Klein, 2000)—can facilitate the integration of members’ unique expertise. Teams with more experiential resources have greater accumulated experience in their particular domain which increases members’ knowledge of pertinent topics, tools, and techniques. When members bring very different areas of expertise to the team, clear and concise descriptions of the problem space will improve the efficiency of communication. Members can draw on past project experiences and know when and how to interject without derailing team discussions. With greater experience comes the ability to adopt learned best practices in ways that improve teams’ understanding of the current project (Gardner et al., 2012).

When team members share little expertise in common, their motivation to share their expertise may suffer. However, with greater experience, it becomes more likely team members have experience working with different expertise domains and can plan for likely challenges. Teams whose members possess more experiential resources may have developed approaches for integrating different expertise and ideas which will improve the collaborative dynamics on the team (Vera & Crossan, 2005). Furthermore, greater experiential resources enable inventors to develop routines for knowledge sharing and a reasonable set of expectations about project work and team functioning. Thus, as the challenges associated with integrating different areas of expertise increase, greater experiential resources on the team can offset them, enhancing teams’ ability to utilize its expertise. Thus,

H3: Team experiential resources positively moderate the relationship between team expertise unsharedness and team innovation such that expertise unsharedness more strongly predicts innovation as team experiential resources increase.

Configuration of Team Relational and Experiential Resources

Although the amount of team relational and experiential resources shape teams’ ability to integrate members’ expertise, so too does the configuration of these resources in the team (Gardner et al., 2012; Kogut & Zander, 1992). Specifically, the distribution versus concentration of these resources across members of the team moderates the effect of expertise unsharedness on innovation. When relational resources are distributed more evenly across members of a team, realizing the benefits of these resources becomes more likely. Put differently, when teams’ relational resources are concentrated in only two or three members, the broader group lacks the common ground needed for integrating different areas of expertise. Furthermore, the concentration of relational resources could lead to the formation of subgroups (cf. Lau & Murnighan, 1998) where efficient communication, the motivation to share expertise, and improved collaborative dynamics occur only among a subset of team members who know and trust each other. In this case, rather than benefiting from the distinct expertise of all members, teams will likely utilize only a portion of their available expertise. Therefore, when relational resources are evenly distributed across members of the team, a team-wide common ground develops where the whole team shares a similar understanding of who knows what, members’ work and interaction preferences, and how to support other members to achieve team goals, ultimately improving interpersonal communications and easing the challenges associated with integrating disparate expertise. Thus,

H4: The distribution of team relational resources positively moderates the relationship between the unsharedness of team expertise and team innovation such that expertise unsharedness more strongly predicts innovation as the distribution of relational resources increases.

Whereas the distribution of relational resources was argued to enhance teams’ ability to leverage the differences in expertise among team members, the concentration of experiential resources is argued to inhibit it. The concentration of experiential resources within a team can create status hierarchies that have negative consequences for efficient communications, information sharing, and collaborative dynamics within a team (Tzabbar & Vestal, 2015). Indeed, several scholars point to status and power as impediments to team creativity, innovation, and performance (Bunderson & Reagans, 2011; Nembhard & Edmondson, 2006). First, status hierarchies limit the open communication of ideas within the team and impede efficient communication. Higher status team members (those with greater relative expertise) are likely to set meeting agendas, dominate team discussions, and talk out of turn (Magee & Galinsky, 2008), ultimately suppressing the contributions of lower status members. With less open exchange, collective learning and expertise integration is undermined, the development of a team-wide common ground is hampered, and the communication routines required for expertise integration fail to develop. Second, much of the literature on the effects of status and power on team innovation suggests status asymmetries decrease the motivation of team members to contribute ideas and stifle team learning behaviors (Bunderson & Reagans, 2011; van der Vegt et al., 2010). Because expert members are more likely to control team resources and to be the interface with external team constituents, lower status members may lack the freedom of self-expression and withhold ideas that run counter to those of the expert members (Bunderson & Reagans, 2011). Finally, the concentration of experiential resources within the team impedes the development of effective collaborative dynamics. Powerful members are less likely to appreciate what other members think and feel and be aware of salient clues about team dynamics (Galinsky et al., 2008; Magee & Galinsky, 2008). In order for teams to effectively integrate member expertise, inventors must feel safe to share information, experiment and take risks, and reflect on their mistakes ((Bunderson & Reagans, 2011; Danneels & Vestal, 2020). In sum, the concentration of experiential resources on the team constrains members’ ability and motivation to engage in meaningful exchange, experiment and take risks, and consider ideas that run counter to established viewpoints (Bunderson & Reagans, 2011), all of which limit teams’ ability to benefit from the differences in expertise available to them. Thus,

H5: The concentration of team experiential resources negatively moderates the relationship between the unsharedness of team expertise and team innovation such that expertise unsharedness has a weaker relationship with innovation as the concentration of experiential resources increases.

Method

The hypotheses are tested using data from research and development teams in the area of nanotechnology. Nanotechnology is a notably highly interdisciplinary field of applied science that draws on physics, chemistry, engineering, materials science, and microbiology (Avenel et al., 2007). The interdisciplinary nature of nanotechnology research requires scientists from these fields to integrate different knowledge domains to achieve team objectives, which makes this an informative context for examining how and when interdisciplinarity relates to team innovativeness.

Publicly available data on firms engaged in nanotechnology research made available through Nanobank. Nanobank is a digital library containing observations from various sources (scientific articles and patents), determined to be related to nanotechnology either by probabilistic information retrieval (IR) methods or by declaration from a source authority (Zucker & Darby, 2009). Nanobank contains data on 240,437 patent applications (that were ultimately granted) from the U.S. Patenting and Trademark Office’s (USPTO) online database from 1946–2004. The data set contains bibliographic information (e.g., inventor names, affiliations, and geo-coding information) from which variables were coded to test the relationships of interest. Additional data were gathered from COMETS to determine firms’ patenting capabilities in technological areas other than nanotechnology (Zucker & Darby, 2011). Finally, to distinguish between inventors with the same or similar names, data from Harvard Business School’s Patent Network Dataverse were used (Lai et al., 2011). In total, this study explored the patenting activities of 49,494 R&D teams conducting research in the area of nanotechnology between 1995 and 2004. It is difficult to trace the origin of nanotechnology as a field of scientific research to a specific date or event. However, it is widely accepted that the discovery of carbon nanotubes in 1991 represents the greatest commercial breakthrough in nanotechnology thus far (Peters, 2015). A substantial increase in patents and scientific publications in nanotechnology followed, punctuated by a greater than 100% increase in patent applications from 1994 to 1995 (Darby & Zucker, 2003). Given this, and the available time frame of data in Nanobank, this study uses 1995 to 2004 as the sample period. For robustness, additional models using multiple different samples starting from 1991 to 1994 were estimated and revealed similar results (available upon request) in terms of the direction and significance of the hypothesized relationships.

Sample construction started with the population of patents assigned to U.S. firms in Nanobank between 1995 and 2004. Next, patents with more than one assignee were dropped from the sample, as these instances are rare and could be the result of unique circumstances such as an R&D alliance. Because the focus of this study is teams, patents with only one inventor were eliminated. The result was a final sample of 49,494 patents. Key study variables are summarized in Table 1.

Table 1.

Variable Definitions and Operationalizations.

Variables	Definition	Operationalization
Dependent
Team innovation	Teams’ ability to generate novel and useful solutions, inventions, or ideas	Number of forward citations to the team’s patent, within 5 years of the patent’s issue date
Independent
Unsharedness of team expertise	The lack of overlap of expertise among team members	The average angular distance between the expertise vectors of the team members
Amount of team relational resources	Composition of the team’s relational resources	Summed count of the number of times each member dyad has co-patented in the past
Amount of team experiential resources	Composition of the team’s experiential resources	Summed count of the number of each team member’s prior patents
Configuration of team relational resources	Compilation (distribution) of the team’s relational resources	Count of the number of times the entire team has co-patented in the past
Configuration of team experiential resources	Compilation (concentration) of the team’s experiential resources	Herfindahl index measure of the concentration of the team members’ prior patents in one or a few members
Controls
Geographic dispersion	The distribution of team members across different geographic locations	Blau index measure of the distribution of team members across geographic locations
Team size	The number of members on the team	A count of the number of inventors listed on the patent
Number of claims	The number of claims made by the patent	A count of the number of claims listed on the patent
Firm age	How long the firm has been engaged in nanotechnology research	A count of the number of years since the firm first patented or published a research article in nanotechnology
Firm patent stock	The firm’s prior patenting capabilities and investments in technology research	A count of the firm’s prior patents in the areas of nanotechnology, computer science, biology, chemistry, semiconductors, science, and engineering
Year fixed effects	Control for unobserved heterogeneity associated with different years	A dummy variable for each patent indicating its application year
Technology fixed effects	Control for unobserved heterogeneity associated with different technologies	A dummy variable for each patent indicating the technology class to which the patent is assigned

Variables

Dependent variable

Team innovation

This study examines the conditions under which interdisciplinarity enhances innovation. Following prior research, this study conceptualizes innovation in interdisciplinary teams as teams’ ability to generate solutions, inventions, or ideas that are not only novel, but also useful (Benoliel & Somech, 2015; West, 2002). In this context, patents provide an objective measure of innovation because of their ability to assess both novelty and usefulness. First, because patents represent important intellectual property rights, patent applications are screened by patent examiners to ensure the application’s claims are both novel and nonobvious (Gittelman, 2008). Thus, for a patent to be granted, the intellectual property it protects must be determined by the USPTO to be novel. Second, data available from the USPTO indicate the number of times a patent is cited in the future (i.e., forward citations), which provides a valid indicator of the patent’s usefulness, an approach which has been widely used in prior research (Fleming, 2001; Fleming & Sorenson, 2004; Trajtenberg, 1990). Forward citations indicate that the cited patents have contributed to future research and that subsequent patents have refined and improved the cited patent (Funk & Owen-Smith, 2017; Trajtenberg, 1990). Thus, innovation is operationalized as the number of forward citations to a focal patent (Lahiri, 2010), within 5 years of the patent’s issue date, minus the number of self-citations. Research suggests that citations to most patents reach their peak within 5 years (Jaffe & Trajtenberg, 2002), but as a robustness check, models are estimated using a measure of innovation with an 8-year citation window, and another measure with no time window, and show similar results. Consistent with prior research, the number of self-citations, defined as the number of times the firm cites its own patents, are subtracted, because this could inflate the number of overall citations (Fleming & Sorenson, 2004).

Independent variable

Team expertise unsharedness

This variable is conceptualized as the sharedness/overlap versus dissimilarity in expertise among team members. As the interdisciplinarity of teams increases, their expertise sharedness decreases, so this variable is termed team expertise “unsharedness” to reflect the extent to which the differences in expertise among members is increasing. Following prior research, the sharedness versus differences in expertise between inventors is measured as the degree of overlap between them (Benner & Waldfogel, 2008). To capture members’ expertise, this study uses each team member’s prior patents. The USPTO classifies patents according to the underlying technology embodied in the patent, and this study uses the technology classes listed in each inventor’s prior patents as a proxy for each inventor’s expertise. For example, an inventor shown to have patented in Technology Class 33 (Biotechnology) is assumed to have expertise in that technological area. Aggregating the technology class data in each inventor’s prior patents allowed for the construction of an expertise vector for each inventor on the team. Specifically, following prior research (Rosenkopf & Almeida, 2003; Tzabbar, 2009), the percentage of assignments to each of the 37 technology classes are used to create a vector that reflects the expertise profile for each inventor. For example, if an inventor has six prior patents where two are assigned to Technology Classification A, one to Technology B, and three assigned to Technology C, the vector ( K = .33, .17, .50) describes the inventor’s expertise profile. Then, the angular distance between each inventor’s expertise vector is calculated. For a team with two inventors, team expertise unsharedness is the angle Ɵ (measured in radians) between vector A and vector B . Specifically, let Inventors A and B have expertise vectors A = A₁i + A₂j + A₃k and B = B₁i + B₂j + B₃k, respectively, where A₁ is the percentage of Inventor A’s patents in technological area i. Then, the angle is

\cos θ = [\frac{A \cdot B}{‖ A ‖ ‖ B ‖}],

where $A \cdot B$ is the dot product of vectors A and B , and $‖ A ‖ ‖ B ‖$ is the magnitude of vector A multiplied by the magnitude of vector B . This measure has a theoretical range from 0 to 1.57. It has a value of zero for teams with members that have identical expertise vectors (complete sharedness), and a value of 1.57 (1.57 radians = 90 degrees) when there is no expertise overlap (complete unsharedness) between members.¹ For teams with more than two members, the average of the pairwise angular distance among the team members is used.

Moderating variables

Amount of team relational resources

This variable measures the amount of prior shared work experience among members of the team. Archival data from each inventor’s prior patents were used to determine the level of relational resources on the team. For each possible dyad on the team, the number of times that dyad has been listed as co-inventors on a patent up to but not including the application year of the focal patent was counted. Then, this count was summed across the possible dyads on the team.

Amount of team experiential resources

Inventors’ amount of prior patents was used as an indicator of their accumulated experience. Each inventor’s prior patents were counted and then summed across all inventors on the team. A higher value indicates a team where the inventors have many prior patents and thus greater task-relevant skills.

Configuration of team relational resources

This variable is conceptualized as the distribution of relational resources across the members of the team. It is operationalized as the count of the number of times the entire inventor team has worked together as co-inventors on a patent up to the year of the focal patent (Hinds et al., 2000). Thus, the measure captures the prior experience of the whole team. For robustness, an additional measure using a well-accepted measure of network density (Borgatti et al., 2013; Reagans et al., 2005) that captures the average amount of prior experience among team members was created. This variable yielded very similar results, but it was highly correlated with other variables and therefore it was not used in the final model. To accurately assign inventors with the same or similar names to the correct patent teams, data made publicly available through Harvard Business School’s Patent Network Dataverse were used (Lai et al., 2011).

Configuration of team experiential resources

This variable is conceptualized as the extent to which task-related experience is concentrated in one or a few team members. Using the patenting histories of each inventor on the team, this measure was operationalized as the degree to which the accumulated prior patents among team members were held by one or a few inventors on the team. The variable was then calculated using the Herfindahl index:

Configuration of experiential resources = [(\sum_{(s = 2)}^{s} I_{i}^{2}) \times (\frac{s}{(s - 1)})],

where I_i refers to the proportion of previous patents belonging to inventor i, relative to the total number of patents belonging to all scientists on the team. Following Hall (2005), a correction term was used to address potential bias resulting from the number of inventors on the team. A higher score indicates one or a few scientists possess the majority of prior patenting experience on the team.

Control variables

Prior research suggests the extent to which team members are spread across multiple geographic locations can enhance innovation (Lahiri, 2010). Utilizing inventor location data from the focal patent, geographic dispersion was operationalized using the Blau index (Lahiri, 2010). In addition, a correction term is included in the calculation to account for potential bias resulting from fewer locations across which inventors can be assigned (Hall, 2005). More formally,

Geographic dispersion = 1 - [(\sum_{(i = 1)}^{n} C_{i}^{2}) \times \frac{n}{(n - 1)}],

where C_i refers to the proportion of team members in location i, and n is the number of locations in which team members are located.

Prior research suggests teams with more scientists have a greater variety of perspectives on which the team can draw to solve problems (Taylor & Greve, 2006). Thus, this study controls for team size, which is operationalized as the number of scientists listed as inventors on the focal patent. Research by Lanjouw and Schankerman (2004) suggests that the number of claims a patent makes is correlated with the number of forward citations a patent receives. Patent claims delineate the scope of the patent, and patents with a broader scope can receive more forward citations. Thus, this study controls for the number of claims on the patent application.

Because organizational age can influence citations to firms’ inventions (Sorensen & Stuart, 2000), firm age is included as a control and is operationalized as the number of years since a firm first patented or published a research article in the area of nanotechnology (Darby & Zucker, 2003). Firm patent stock controls for a firm’s prior patenting capabilities and was operationalized as the count of the total number of patents in the technological areas of nanotechnology, computer science, biology, chemistry, semiconductors, other science, and other engineering. These data were gathered from COMETS (Zucker & Darby, 2011). Finally, year fixed effects and technology class fixed effects are included in all models to control for sources of unobserved heterogeneity.

Results

Model Estimation and Statistical Interpretation

The dependent variable (innovation) is a count-type variable and takes only positive integer values. In this study, these data are overdispersed and thus violate the Poisson regression assumption of the variance equal to the mean. Therefore, negative binomial regression models that estimate an additional parameter that directly models the overdispersion were used (Cameron & Trivedi, 2010). A cluster-robust estimator to produce standard errors that account for the potential correlation of error terms between patents produced by the same firm is also calculated (Cameron & Trivedi, 2010).

Negative binomial models are nonlinear so the interpretation of model coefficients is not straightforward (Zelner, 2009). The direction, magnitude, and significance of the interaction effect in nonlinear models can vary for different values of the covariates and are not directly interpretable from the model coefficients and standard errors (Ai & Norton, 2003). Thus, a simulation approach developed by Zelner (2009), with the STATA command INTGPH, is used to examine the sign and significance of the moderation effect over the entire range of the independent variable (Zelner, 2009).

Findings

Table 2 reports the descriptive statistics and correlations for all model variables. The average team had an expertise unsharedness score of 0.48 radians (28 degrees) and generated patents receiving 9.95 forward citations. Teams on average had 3.56 inventors, 17.61 prior collaborations, and a stock of 69.23 patents. Results show a positive correlation between team member expertise unsharedness and innovation. Results also suggest larger teams are associated with greater innovation. Team size is correlated at .70 with the concentration of experiential resources, due in part to the bias correction term used to account for the number of inventors across which knowledge resources on the team are assigned (i.e., team size). Nonetheless, after centering the model variables, the variance inflation factors (VIFs) show an average VIF of 1.71, and the highest model VIF was 5.18 which is below the benchmark of 10.0, suggesting multicollinearity was not a problem. Furthermore, Allison (2012) suggests multicollinearity resulting from the inclusion of product terms does not have adverse consequences.

Table 2.

Descriptive Statistics and Correlations.

Variable	M	SD	1	2	3	4	5	6	7	8	9	10
1. Team innovation	9.95	18.05
2. Team expertise unsharedness	0.48	0.36	.06*
3. Amount of team relational resources	17.61	59.21	−.03*	−.11*
4. Amount of team experiential resources	69.23	103.34	−.04*	−.14*	.66*
5. Distribution of team relational resources	0.72	3.06	−.02*	−.13*	.27*	.34*
6. Concentration of team experiential resources	0.79	0.42	−.03*	.09*	−.24*	−.20*	.04*
7. Geographic dispersion	0.31	0.42	.02*	.04*	.00	−.02*	.01	−.14*
8. Team size	3.56	1.89	.04*	−.00	.36*	.31*	−.08*	−.70*	.15*
9. Number of claims	25.11	21.79	.13*	.02*	.00	.02*	.00	−.03*	.03*	.08*
10. Firm age^a	2.32	1.10	−.12*	.03*	.07*	.14*	.01*	−.06*	−.02*	.08*	−.12*
11. Firm patent stock	502.51	777.25	−.06*	.06*	.08*	.19*	.03*	−.07*	−.02*	.07*	−.08*	.54*

Log.

p < .05.

The results of the negative binomial regression analyses predicting innovation are summarized in the three models reported in Table 3. Model 1 shows the effects of the control and moderating variables. Model 2 includes a test of the effects of team expertise unsharedness on innovation, and Model 3 is the fully specified model. Models 2 to 3 support H1, suggesting team expertise unsharedness has an inverted U relationship with innovation. Results of Model 3 show the coefficient on team expertise unsharedness is positive and significant (0.45, p < .001), and the coefficient on the squared term of team expertise unsharedness is negative and significant (−0.20, p< .05). Using the coefficients from Model 2, the inflection point at which increasing levels of expertise unsharedness (i.e., less overlap) has a negative effect on innovation is calculated to be 0.59 radians. This relationship is depicted in Figure 2. Finally, because there are observations in the sample with expertise unsharedness scores on both sides of the optimum point (approximately 3,500 teams are beyond the optimum point), the results provide strong support for H1 (Aiken & West, 1991).

Table 3.

Negative Binomial Regression Analysis Predicting Team Innovation.

Variable	Model	Model	Model
Variable	(1)	(2)	(3)
Team expertise unsharedness		0.40***	0.45***
Team expertise unsharedness		(0.05)	(0.06)
Team expertise unsharedness squared		−0.34***	−0.20*
Team expertise unsharedness squared		(0.09)	(0.08)
Team Expertise Unsharedness × Amount of Team Relational Resources^a			−0.13
			(0.11)
Team Expertise Unsharedness × Amount of Team Experiential Resources^a			0.26**
			(0.08)
Team Expertise Unsharedness × Distribution of Team Relational Resources			0.08**
			(0.03)
Team Expertise Unsharedness × Concentration of Team Experiential Resources			−0.15*
			(0.06)
Amount of team relational resources^b	−0.04	0.06	−0.24
Amount of team relational resources^b	(0.40)	(0.38)	(0.31)
Amount of team experiential resources^b	−0.65	−0.58	−0.03
	(0.45)	(0.42)	(0.35)
Distribution of team relational resources^a	−0.48	−0.21	2.36*
Distribution of team relational resources^a	(0.36)	(0.37)	(0.96)
Concentration of team experiential resources^a	0.35	−2.19	−1.74
Concentration of team experiential resources^a	(4.20)	(4.03)	(3.78)
Geographic dispersion	0.04	0.03	0.03
Geographic dispersion	(0.03)	(0.03)	(0.03)
Team size	0.06***	0.04***	0.04***
Team size	(0.01)	(0.01)	(0.01)
Number of claims	0.01***	0.01***	0.01***
Number of claims	(0.00)	(0.00)	(0.00)
Firm age^c	−0.13***	−0.13***	−0.13***
Firm age^c	(0.02)	(0.02)	(0.02)
Firm patent stock^b	−0.04	−0.04	−0.04
Firm patent stock^b	(0.05)	(0.05)	(0.05)
Constant	1.46*** (0.12)	1.47*** (0.12)	1.49*** (0.12)
Year fixed effects	Y	Y	Y
Technology fixed effects	Y	Y	Y
Log pseudo likelihood	−156,716.4	−156,485.6	−156,359.34
Wald χ²	3,030.60	3,283.63	3,139.67
N	49,494	49,494	49,494

Note. Standard errors are in parentheses.

Multiplied by 10² for interpretation purposes. ^b Multiplied by 10³ for interpretation purposes. ^c log.

p < .05. **p < .01. ***p < .001.

Figure 2.

The effect of team expertise unsharedness on innovation.

As this correlation is nonlinear, it is prudent to explore the relationship between team expertise unsharedness and innovation at different points along the inverted U. Using the coefficients in Model 2, results show that along the positive slope of the curve, increasing the unsharedness of team member expertise from low levels (mean minus one standard deviation = 0.13 radians) to the optimum level (0.59 radians is the inflection point) increases innovation by approximately 7%. Similarly, exploring the negative slope of the curve, results show that a one standard deviation increase in the unsharedness of team members’ expertise from the optimum level results in an approximately 4% decrease in innovation.²

Importantly, as discussed above, the sign and significance of interaction terms in a negative binomial model cannot be interpreted solely from the model coefficients and standard errors (Zelner, 2009). To this end, the results are presented graphically (Hoetker, 2007; Zelner, 2009) in Figures 3 to 5. Figures 3A, 4A, and 5A reveal the sign and significance of the interactions at different values of the covariates (i.e., a change from mean to mean plus one standard deviation in the moderator over the range of expertise unsharedness). The markers on the plotted lines depict the range of values for which the coefficient of the interaction term is significantly different from zero at the 95% confidence level (Zelner, 2009). Figures 3B, 4B, and 5B depict the magnitude of the effects and show the level of innovation associated with different levels of expertise unsharedness at low (mean minus one standard deviation) versus high (mean plus one standard deviation) levels.³

Figure 3.

(A) The moderating effect of the amount of team experiential resources on the team expertise unsharedness to innovation relationship. (B) The moderating effect of the amount of team experiential resources on the team expertise unsharedness to innovation relationship (values of team expertise unsharedness are mean centered).

Figure 4.

(A) The moderating effect of the distribution of team relational resources on the team expertise unsharedness to innovation relationship. (B) The moderating effect of the distribution of team relational resources on the team expertise unsharedness to innovation relationship (values of team expertise unsharedness are mean centered).

Figure 5.

(A) The moderating effect of the concentration of team experiential resources on the team expertise unsharedness to innovation relationship. (B) The moderating effect of the concentration of team experiential resources on the team expertise unsharedness to innovation relationship (values of team expertise unsharedness are mean centered).

Amount of relational and experiential resources

H2 argued that the amount of relational resources helps to build common ground and shared understandings that help teams integrate the differences in expertise available to them. Contrary to predictions, the coefficient on the interaction term was not significant (p > .05); thus, H2 is not supported. Potential reasons for this are discussed later.

H3 predicted that higher amounts of team experiential resources would increase inventor’s understanding of the problem space as well as their ability to utilize different perspectives, and thus positively moderate the relationship between team expertise unsharedness and innovation. Results in Model 3 show the coefficient is positive and highly significant (0.003, p < .001), providing support for H3. Figure 3A, which indicates the sign and significance of the interaction, shows an interesting pattern of results. First, as predicted, higher levels of team experiential resources moderate the unsharedness to innovation relationship. The plot shows the moderating effect is positive and significant only as the unsharedness in expertise among team members approaches high levels (> 0.84). However, at low levels of unsharedness, team experiential resources have a negative and significant moderating effect. Perhaps when inventors share similar knowledge, high levels of accumulated experience in similar domains prevent inventors from exploring new ideas (e.g., a familiarity trap).

To further probe the nature of the interaction, the marginal effects at meaningful values of the independent and moderator variables are depicted graphically (Hoetker, 2007). Figure 3B shows the effect of team expertise unsharedness on innovation at low (mean minus one standard deviation) and high (mean plus one standard deviation) levels of experiential resources. Using the coefficients in Model 3, results show that at high levels of expertise unsharedness (mean plus one standard deviation), a change in the level of experiential resources from low to high levels results in an approximately 21% increase in innovation.

Configuration of relational and experiential resources

H4 predicted that the distribution of team relational resources would facilitate the development of a team-wide common ground and shared understanding, and thus would have a positive moderating effect on the team expertise unsharedness–innovation relationship. Consistent with our hypothesis, the coefficient in Model 3 is positive and significant (0.08, p < .01). Figure 4A shows sign and significance of the interaction over the range of values of expertise unsharedness, indicating relational resources have a significant and positive moderating effect at levels of expertise unsharedness slightly below the mean level, and beyond. Interestingly, Figure 4A shows a significant and negative moderating effect at low levels of expertise unsharedness. This suggests that when the expertise among inventors is largely redundant, dense intrateam ties close the team off to outside ideas and thus limits their innovative potential.

In Figure 4B, and similar to above, the effect of team expertise unsharedness at low and high values of the distribution of relational resources is depicted graphically. Using the coefficients in Model 3, results reveal that at high levels of expertise unsharedness, an increase in the distribution of relational resources from low to high levels results in an approximately 37% increase in innovation.

Finally, H5 argued that the concentration of experiential resources on the team creates power hierarchies that limit teams’ abilities to integrate member expertise and thus negatively moderate the expertise unsharedness–innovation relationship. As predicted, the coefficient on the interaction term is negative and significant (−0.15, p < .05). Figure 5A depicts the sign and significance of this interaction over the range of values of expertise unsharedness. It reveals that at low values of expertise unsharedness, no significant interaction is found. However, when the unsharedness among members’ expertise is greater than 0.65 radians, the concentration of experiential resources on the team has a negative moderating effect on teams innovation.

In Figure 5B, and similar to above, the effect of team expertise unsharedness on innovation at low and high levels of the concentration of experiential resources is depicted graphically. At high levels of expertise unsharedness, an increase in the concentration of experiential resources from low to high levels decreases innovation by approximately 6%.

Overall, these results provide support for arguments that the differences in expertise between inventors on a team have an inverted U relationship with innovation. Inventors benefit from their distinct expertise up to a point, after which increasing unsharedness in their expertise has a deleterious effect on innovation. Results show that while teams with greater experiential resources are better able to utilize members’ unique expertise, teams wherein experiential resources are concentrated in one or a few members are not. Interestingly, whereas the even distribution of relational resources on the team has a positive moderating effect on the unsharedness to innovation relationship, no evidence is found that the amount of relational resources on the team shapes teams’ ability to benefit from members’ unique expertise. Perhaps the configuration (in this case the even distribution) of relational resources is considerably more important for knowledge integration than is the level of relational resources. These results provide support for the idea that the amount and configuration of relational and experiential resources on the team are an important determinant in understanding when interdisciplinary team members are able to leverage their unique areas of expertise to enhance innovation.

Discussion

The promise of teams lies in their potential to bring together different ideas and perspectives to generate higher quality solutions than would be possible for individuals working alone. Teams experts continue to explore the factors that differentiate teams who are able to successfully integrate their unique insights from those who are not, and have identified factors associated with team cognition and processes (e.g., Hülsheger et al., 2009; Mathieu et al., 2000) as well as communication media (e.g., Mesmer-Magnus et al., 2011; K. Zhang & Ge, 2006) and technological supports and constraints (e.g., Riebe et al., 2016) that can either help or hinder team performance. Using data from over 49,000 R&D teams in the nanotechnology sector, this study explored how the amount and configuration of team experiential and relational resources interact to determine when team expertise unsharedness will promote team innovativeness. Results suggest teams composed of members from a variety of disciplines (interdisciplinary teams) may have greater “potential” expertise, but not all of the team’s expertise can be leveraged for innovation and creativity. Rather, a team’s “instrumental” expertise—that expertise which can be leveraged by the team—is a function of both the amount and configuration of the team’s experiential and relational resources.

Capitalizing on Team Interdisciplinarity: Potential Versus Instrumental Expertise

Often teams are intentionally composed of members from different disciplines with the goal of increasing their creativity and innovation (Ahmadpoor & Jones, 2019; Salazar et al., 2012). Unfortunately, in practice, these teams frequently fail to reach their innovative potential, often because they are not able to integrate their unique expertise into a unified whole (Gardner et al., 2012; Xiao et al., 2016). The results suggest the differences in expertise among team members enhance innovation up to a point, after which those differences hurt innovation. In other words, a team’s instrumental expertise may be less than their potential expertise, wherein the difference between potential and instrumental expertise is a function of both the amount and configuration of the team’s relational and experiential resources. Logically, it would seem that, for both relational and experiential resources, more should be better. It is argued that shared team experience improves teams’ ability to leverage member expertise because it facilitates the development of familiarity (Espinosa et al., 2007), trust (Uzzi & Lancaster, 2003), common ground (Cramton, 2001), and efficient team processes (Mathieu et al., 2000). Furthermore, the more task-relevant experience a team has to draw on, the greater the chances the team can leverage member expertise and develop truly innovative ideas. Greater experience increases members’ knowledge of relevant tools and techniques for working across disciplines. More experienced members also know when and how to interject with disrupting communications and have likely developed an awareness of other disciplines, something less experienced members would lack.

However, by applying research from the RBV and combinative capabilities of teams (Gardner et al., 2012), it is argued that simply considering the total amount of these resources within a team is insufficient to predict the team’s instrumental expertise. Rather, it is the amount of these resources along with how they are configured (i.e., distributed vs. concentrated) within a team that determines its ability to capitalize on potential expertise. In particular, results suggest that teams wherein relational resources are evenly distributed among members are better able to leverage member expertise to enhance innovation. Moreover, whereas results show teams with more experiential resources are better equipped to leverage member expertise, teams wherein experience is concentrated in one or a few members struggle to utilize their potential. Thus, this study offers a framework for exploring the conditions under which interdisciplinary teams can overcome what research purports is their greatest obstacle—integrating member expertise (Fiore, 2008; Salazar et al., 2012; van Knippenberg, 2017).

Interdisciplinarity Is a Matter of Degree

The use of interdisciplinary teams and research on team interdisciplinarity is growing in popularity, although most of this research has been conceptual or has tested hypotheses in the context of interdisciplinary teams rather than exploring the mechanisms of interdisciplinarity that make these teams maximally effective (cf. Benoliel & Somech, 2015; Salazar et al., 2012; Yong et al., 2014). Importantly, team interdisciplinarity is a matter of degree, and conceptualizing it as such can open a new line of inquiry in the study of interdisciplinary teams. This study puts forth an alternative conceptualization and measure of interdisciplinarity, based on our current understanding of the strengths and weaknesses of interdisciplinary teams, that can move the conversation forward. In particular, a key takeaway of these findings is that the common approach for conceptualizing and operationalizing team expertise as diversity or variety of member knowledge (Harrison & Klein, 2007), or as the spread of members’ expertise across expertise domains, although insightful, leaves important variance unexplained. That is, it ignores the breadth of individual members’ expertise and does not accurately capture the extent to which team members differ in their expertise (Bunderson & Sutcliffe, 2002). This is an important point given that the primary strength of interdisciplinary teams is thought to be a function of the differences in disciplinary expertise among team members (Salazar et al., 2012).

Limitations and Directions for Future Research

As with any study, this research has limitations which should be considered when interpreting the results. First, due to the archival nature of the data, this study was not able to (a) directly observe knowledge sharing and integration among inventors, or (b) model the motivation to form an interdisciplinary team nor the motivation for choosing particular inventors. Although this study attempts to control for sources of unobserved heterogeneity by controlling for year and technology fixed effects that may allay these concerns, future research is needed to isolate optimal knowledge sharing processes for interdisciplinary teams, particularly at varying degrees of interdisciplinarity. Another potential limitation of this research relates to the generalizability of the findings to other contexts. This study focused on a high technology setting wherein teams comprised highly skilled innovators. Future research is needed to explore the extent to which the aspects of interdisciplinarity studied here are relevant in other contexts, particularly within simpler task environments.

A third potential limitation of this research relates to the nature of the operationalization of amount and configuration of relational and experiential resources. Due to the archival nature of the data, this study used the number of times dyads within the total team had worked together on prior patents as proxies for the amount and configuration of relational resources on the team. Similarly, this study used the sum and the concentration of prior patents belonging to inventors on the team as proxies for the amount and concentration of experiential resources on the team. Although the archival data used herein overcome limitations associated with perceptual measures of knowledge integration and performance, these operationalizations of relational resources do not necessarily assess emergent states (e.g., transactive memory) or team processes (e.g., communication or coordination norms) that may have developed within teams. Counts of prior patents as evidence of experiential resources do not necessarily assess the extent to which the inventors were actively or equivalently involved in prior projects and thus serve as an imperfect proxy for the nature and degree of team member experience. Arguably, this is less of a concern for teams in this context given the keen attention paid to intellectual property and commercialization rights associated with inventors listed on patents, but could be of concern in other teamwork contexts. Furthermore, the measures of relational resources do not account for any familiarity among members that may have occurred outside of patent projects (e.g., other research co-authorships or social interactions). Similarly, because some inventors may work remotely or in other geographic locations, it is possible that not all co-inventors listed on the patent gained equal amounts of familiarity with the other members on the team.

This study’s research design and results suggest several profitable directions for future research. First, future research should directly examine the effects of the amount and configuration of team experiential and relational resources on the mechanisms described in the current study such as shared mental models, transactive memory systems, and collaborative dynamics. Second, research is needed to explore how to help teams harness their potential expertise, particularly in situations wherein their degree of interdisciplinarity is suboptimal. Research related to the development and maintenance of shared mental models (Mathieu et al., 2000; Mohammed et al., 2000), transactive memory (Hollingshead & Brandon, 2003; Lewis, 2003), and team information sharing (Mesmer-Magnus & DeChurch, 2009; Stasser et al., 2000) would be relevant here.

This study’s results also suggest scholars’ choice between expertise variety or unsharedness constitutes an important research design decision that can influence empirical results and how those results should be interpreted (Bunderson, 2003; Harrison & Klein, 2007). The findings would suggest team innovative potential is a complex interplay between the differences in expertise among members and amount of team familiarity and experiences as well as how those resources are concentrated versus distributed within the team. Whereas prior research has examined the direct effect of team resources on knowledge integration (cf. Gardner et al., 2012), this study puts forth a moderation model that examines how team resources condition teams’ ability to utilize member expertise to enhance innovation, shedding new light on how team inputs and processes interact. Furthermore, by developing a research model highlighting the level and structure of resources on the team, this study draws a connection between an RBV of the team and the team’s social structure, underscoring the importance of teams’ social structures that can either facilitate or impair knowledge sharing and integration and adding to a growing body of research focused on understanding knowledge integration in teams (Gardner et al., 2012; Salazar et al., 2012; Taylor & Greve, 2006).

Finally, the results did not support the hypothesis regarding the amount of relational resources as a moderator of the team expertise–team innovation relationship. It may be that the compilation of team familiarity within the team is more critical than the amount. Alternatively, there may be another moderator at play. For example, perhaps the nature of prior interactions/processes affects the extent to which familiarity supports expertise integration. Network theory (Katz et al., 2004) might offer a useful framework and methodology for studying how the amount and configuration of relational resources affects team innovative potential. Network (Katz et al., 2004) and vertical dyad (Schriesheim et al., 1999) theories, as well as research on team faultlines (Rico et al., 2007; van Knippenberg et al., 2011), would suggest concentrations of familiarity within certain dyads of a system/team could promote gatekeeping, ingroup/outgroup dynamics, as well as other inefficient processes, such as members of the team being left out of important communications, relevant member knowledge not being accessed, bottlenecks in coordination or decision-making.

Practical Implications

From a practical standpoint, these findings suggest important team design strategies for practitioners. This study demonstrates that simply comprising a team of individuals from diverse disciplines does not necessarily enhance innovativeness. A team comprised members with very different expertise but not enough shared experience working together on prior projects may be at a disadvantage compared with a team wherein the members have both unique expertise and shared experience. On the contrary, although expertise similarities likely facilitate efficient communication and workflows, such teams’ outputs are likely to be more incremental in nature. As such, managers should deploy teams wherein members have moderate differences in their expertise such that expertise and familiarity are distributed across the team, so members have sufficient common understanding and motivation to be able to integrate their unique expertise.

Second, when project requirements dictate that teams be composed of members with very different bases of expertise, it would be prudent to bring together individuals with high, but relatively equal amounts of prior experience as well as individuals who have worked together as a team in the past. Teams wherein experience and familiarity are concentrated within a subset of the team may not be able to capitalize on the expertise of other team members. For example, in research domains like nanotechnology, expert inventors tend to control the team’s research agenda and resources which can increase the power of the expert inventor(s) and inhibit the contributions of other members (Huang & Cummings, 2011). In addition, familiarity among members of a team should be distributed across subsets of the team rather than concentrated within a few members. However, it may not be feasible or possible for managers to form teams wherein members have a high degree of familiarity with each other. In such cases, particular attention is needed to help the team quickly build shared cognitive architectures so the members of the team are able to leverage their distributed expertise (Healey et al., 2015). For example, interventions aimed at developing transactive memory within the team will help team members be on the same page as to who knows what and who can be relied upon for what in their team (Lewis, 2003). Finally, research would suggest that efforts to quickly build team identity and effective team processes are also likely to be particularly important for innovation teams who cannot be comprised on the basis of familiarity and expertise (Bell et al., 2018).

Third, it is important to assess the nature of shared cognition and team processes within interdisciplinary teams to enhance the compilational expertise among members as well as to put in place countermeasures before breakdowns in communications yield detrimental implications for team performance. Managers must be aware that simply designing an interdisciplinary team does not in and of itself guarantee maximal performance. This study’s results demonstrate a clear difference between a team’s potential expertise and its ability to leverage it.

Conclusion

While the differences in expertise possessed by team members offer the potential for valuable inventions, teams face challenges integrating their knowledge effectively (cf. Salazar et al., 2012). Teams experts have continued to ask, Why are some teams better able to translate distinct member expertise into higher quality team outcomes than others? This study explored this question using inventors in R&D teams within the nanotechnology sector as a backdrop, and investigated whether and under what conditions team expertise unsharedness promotes team innovativeness. Results showed a complex relationship between expertise unsharedness among team inventors and innovation, confirming predictions that expertise unsharedness within a team has an inverted-U relationship with innovation (cf. Dougherty, 1992; Fleming, 2001), such that moderate degrees of unsharedness in experience and familiarity reflect optimal innovative potential. This study argued that research on combinative capabilities and the RBV can be extended to the team level to provide a useful framework for understanding team innovation. In sum, team experiential and relational resources, because they can facilitate or impede the integration of members’ distinct expertise, are important for understanding when teams are able to most effectively leverage their interdisciplinarity.

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) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Alex Vestal

Notes

Author Biographies

Alex Vestal is an assistant professor of management at the University of North Carolina Wilmington, USA. He earned his PhD from the University of Central Florida in 2011. His research focuses on knowledge sharing as a facilitator of innovation. His research has been published in Organization Science and Research Policy.

Jessica Mesmer-Magnus is a professor of management in the Cameron School of Business at the University of North Carolina Wilmington, USA. She earned her PhD in organizational psychology at Florida International University in 2005. Her research focuses in the area of teams, and has been funded by ARI and NASA.

References

Ahmadpoor

Jones

B. F.

(2019). Decoding team and individual impact in science and invention. Proceedings of the National Academy of Sciences, 116(28), 13885–13890. https://doi.org/10.1073/pnas.1812341116

Norton

E. C.

(2003). Interaction terms in logit and probit models. Economic Letters, 80(1), 123–129. https://doi.org/10.1016/S0165-1765(03)00032-6

Aiken

L. S.

West

S. G.

(1991). Multiple regression: Testing and interpreting interactions. Sage.

Allison

(2012). When can you safely ignore multicollinearity. Statistical Horizons, 5(1). https://statisticalhorizons.com/multicollinearity

Avenel

Favier

A. V.

Magematin

Rieu

(2007). Diversification and hybridization in firm knowledge bases in nanotechnologies. Research Policy, 36(6), 864–870. https://doi.org/10.1016/j.respol.2007.02.002

Baumann

M. R.

Bonner

B. L.

(2013). Member awareness of expertise, information sharing, information weighting, and group decision making. Small Group Research, 44(5), 532–562. https://doi.org/10.1177/1046496413494415

Beck

S. J.

Meinecke

A. L.

Matsuyama

Lee

C. C.

(2017). Initiating and maintaining collaborations and facilitating understanding in interdisciplinary group research. Small Group Research, 48(5), 532–543. https://doi.org/10.1177/1046496417721746

Bell

S. T.

Brown

S. G.

Weiss

J. A.

(2018). A conceptual framework for leveraging team composition decisions to build human capital. Human Resource Management Review, 28(4), 450–463. https://doi.org/10.1016/j.hrmr.2017.06.003

Bell

S. T.

Villado

A. J.

Lukasik

M. A.

Belau

Briggs

A. L.

(2011). Getting specific about demographic diversity variable and team performance relationships: A meta-analysis. Journal of Management, 37(3), 709–743. https://doi.org/10.1177/0149206310365001

10.

Benner

Waldfogel

(2008). Close to you? Bias and precision in patent-based measures of technological proximity. Research Policy, 37(9), 1556–1567. https://doi.org/10.1016/j.respol.2008.05.011

11.

Benoliel

Somech

(2015). The role of leader boundary activities in enhancing interdisciplinary team effectiveness. Small Group Research, 46(1), 83–124. https://doi.org/10.1177/1046496414560028

12.

Borgatti

S. P.

Everett

M. G.

Johnson

J. C.

(2013). Analyzing social networks. SAGE.

13.

Bunderson

J. S.

(2003). Recognizing and utilizing expertise in work groups: A status characteristics perspective. Administrative Science Quarterly, 48(4), 557–591. https://doi.org/10.2307/3556637

14.

Bunderson

J. S.

Reagans

R. E.

(2011). Power, status, and learning in organizations. Organization Science, 30, 1–13. https://doi.org/10.1287/orsc.1100.0590

15.

Bunderson

J. S.

Sutcliffe

K. M.

(2002). Comparing alternative conceptualizations of functional diversity in management teams: Process and performance effects. Academy of Management Journal, 45(5), 875–893. https://doi.org/10.5465/3069319

16.

Cameron

Trivedi

R. E.

(2010). Microeconometrics using Stata. Stata Press.

17.

Carlile

P. R.

(2004). Transferring, translating, and transforming: An integrative framework for managing knowledge across boundaries. Organization Science, 15(5), 555–568. https://doi.org/10.1287/orsc.1040.0094

18.

Cohen

W. M.

Levinthal

D. A.

(1990). Absorptive capacity: A new perspective on learning and innovation. Administrative Science Quarterly, 35(1), 128–152. https://doi.org/10.2307/2393553

19.

Cramton

C. D.

(2001). The mutual knowledge problem and its consequences for dispersed collaboration. Organization Science, 12(3), 346–337. https://doi.org/10.1287/orsc.12.3.346.10098

20.

Cronin

M. A.

Weingart

L. R.

(2007). Representational gaps, information processing, and conflict in functionally diverse teams. Academy of Management Review, 32(3), 761–773. https://doi.org/10.5465/amr.2007.25275511

21.

Danneels

Vestal

(2020). Normalizing vs. analyzing: Drawing the lessons from failure to enhance firm innovativeness. Journal of Business Venturing, 35(1), 1–18. https://doi.org/10.1016/j.jbusvent.2018.10.001

22.

Darby

M. R.

Zucker

L. G.

(2003). Grilichesian breakthroughs: Inventions of methods of inventing and firm entry in nanotechnology (No. w9825). National Bureau of Economic Research.

23.

DeChurch

L. A.

Mesmer-Magnus

J. R.

(2010). The cognitive underpinnings of effective teamwork: A meta-analysis. Journal of Applied Psychology, 95(1), 32–53. https://doi.org/10.1037/a0017328

24.

De Dreu

C. K

. (2007). Cooperative outcome interdependence, task reflexivity, and team effectiveness: A motivated information processing perspective. Journal of Applied Psychology, 92(3), 628–638. https://doi.org/10.1037/0021-9010.92.3.628

25.

Dougherty

(1992). Interpretive barriers to successful product innovation in large firms. Organization Science, 3(2), 179–202. https://doi.org/10.1287/orsc.3.2.179

26.

Edmondson

A. C.

(1999). Psychological safety and learning behavior in work teams. Administrative Science Quarterly, 44, 350–383. https://doi.org/10.2307/2666999

27.

Edmondson

A. C.

(2002). The local and variegated nature of learning in organization: A group-level perspective. Organization Science, 13(2), 128–146. https://doi.org/10.1287/orsc.13.2.128.530

28.

Edmondson

A. C.

(2012). Teaming: How organizations learn, innovate, and compete in the knowledge economy. John Wiley.

29.

Edmondson

A. C.

Nembhard

I. M.

(2009). Product development and learning in project teams: The challenges are the benefits. Journal of Product Innovation Management, 26(2), 123–138. https://doi.org/10.1111/j.1540-5885.2009.00341.x

30.

Espinosa

J. A.

Slaughter

S. A.

Kraut

R. E.

Herbsleb

J. D.

(2007). Familiarity, complexity, and team performance in geographically distributed software development. Organization Science, 18(4), 613–630. https://doi.org/10.1287/orsc.1070.0297

31.

Fiore

S. M.

(2008). Interdisciplinarity as teamwork: How the science of teams can inform team science. Small Group Research, 39(3), 251–277. https://doi.org/10.1177/1046496408317797

32.

Fleming

(2001). Recombinant uncertainty in technological search. Management Science, 47(1), 117–132. https://doi.org/10.1287/mnsc.47.1.117.10671

33.

Fleming

Sorenson

(2004). Science as a map in technological search. Strategic Management Journal, 25(8–9), 909–928. https://doi.org/10.1002/smj.384

34.

Funk

R. J.

Owen-Smith

(2017). A dynamic network measure of technological change. Management Science, 63(3), 791–817. https://doi.org/10.1287/mnsc.2015.2366

35.

Galinsky

A. D.

Magee

J. C.

Gruenfeld

D. H.

Whitson

J. A.

Liljenquist

K. A.

(2008). Power reduces the press of the situation: Implications for creativity, conformity, and dissonance. Journal of Personality and Social Psychology, 95(6), 1450–1466. https://dx-doi-org.web.bisu.edu.cn/10.1037/a0012633

36.

Gardner

H. K.

Gino

Staats

B. R.

(2012). Dynamically integrating knowledge in teams: Transforming resources into performance. Academy of Management Journal, 55(4), 998–1022. https://doi.org/10.5465/amj.2010.0604

37.

Gittelman

(2008). A note on the value of patents as indicators of innovation: Implications for management research. Academy of Management Perspectives, 22(3), 21–27. https://doi.org/10.5465/amp.2008.34587992

38.

Grant

R. M.

(1996). Toward a knowledge-based theory of the firm. Strategic Management Journal, 17(Suppl. 2), 109–122. https://doi.org/10.1002/smj.4250171110

39.

Hall

B. H.

(2005). A note on the bias in Herfindahl-type measures based on count data. Revue d’économie industrielle, 110(1), 149–156. https://doi.org/10.3406/rei.2005.3076

40.

Harrison

D. A.

Klein

K. J.

(2007). What’s the difference? Diversity constructs as separation, variety, or disparity in organizations. Academy of Management Review, 32(4), 1199–1228. https://doi.org/10.5465/amr.2007.26586096

41.

Harrison

D. A.

Price

K. H.

Gavin

J. H.

Florey

A. T.

(2002). Time, teams, and task performance: Changing effects of surface-and deep-level diversity on group functioning. Academy of Management Journal, 45(5), 1029–1045. https://doi.org/10.5465/3069328

42.

Healey

M. P.

Vuori

Hodgkinson

G. P.

(2015). When teams agree while disagreeing: Reflexion and reflection in shared cognition. Academy of Management Review, 40(3), 399–422. https://doi.org/10.5465/amr.2013.0154

43.

Hinds

P. J.

Carley

K. M.

Krackhardt

Wholey

(2000). Choosing work group members: Balancing similarity, competence, and familiarity. Organizational Behavior and Human Decision Processes, 81(2), 226–251. https://doi.org/10.1006/obhd.1999.2875

44.

Hoetker

(2007). The use of logit and probit models in strategic management research: Critical issues. Strategic Management Journal, 28(4), 331–343. https://doi.org/10.1002/smj.582

45.

Hoever

I. J.

Zhou

van Knippenberg

(2018). Different strokes for different teams: The contingent effects of positive and negative feedback on the creativity of informationally homogeneous and diverse teams. Academy of Management Journal, 61(6), 2159–2181. https://doi.org/10.5465/amj.2016.0642

46.

Hollingshead

A. B.

Brandon

D. P.

(2003). Potential benefits of communication in transactive memory systems. Human Communication Research, 29(4), 607–615. https://doi.org/10.1111/j.1468-2958.2003.tb00859.x

47.

Huang

Cummings

J. N.

(2011). When critical knowledge is most critical: Centralization in knowledge-intensive teams. Small Group Research, 42(6), 669–699. https://doi.org/10.1177/1046496411410073

48.

Huckman

R. S.

Staats

B. R.

Upton

D. M.

(2009). Team familiarity, role experience, and performance: Evidence from Indian software services. Management Science, 55(1), 85–100. https://doi.org/10.1287/mnsc.1080.0921

49.

Hülsheger

U. R.

Anderson

Salgado

J. F.

(2009). Team-level predictors of innovation at work: A comprehensive meta-analysis spanning three decades of research. Journal of Applied Psychology, 94(5), 1128–1145. https://doi.org/10.1037/a0015978

50.

Jaffe

A. B.

(1986). Technological opportunity and spillovers of R&D: Evidence from firms’ patents, profits, and market value. American Economic Review, 76(5), 984–1001. https://doi.org/10.3386/w1815

51.

Jaffe

A. B.

Trajtenberg

(2002). Patents, citations, and innovations: A window on the knowledge economy. MIT press.

52.

Katz

Lazer

Arrow

Contractor

(2004). Network theory and small groups. Small Group Research, 35(3), 307–332. https://doi.org/10.1177/1046496404264941

53.

Keyton

Heylen

D. K.

(2017). Pushing interdisciplinary in the study of groups and teams. Small Group Research, 48(5), 621–630. https://doi.org/10.1177/1046496417732528

54.

Kogut

Zander

(1992). Knowledge of the firm, combinative capabilities, and replication of technology. Organization Science, 3(3), 383–397. https://doi.org/10.1287/orsc.3.3.383

55.

Kozlowski

S. W. J.

Klein

K. J.

(2000). A multilevel approach to theory and research in organizations: Contextual, temporal, and emergent processes. In Klein

K. J.

Kozlowski

S. W. J.

(Eds.), Multilevel theory, research, and methods in organizations: Foundations, extensions, and new directions (pp. 3–90). John Wiley.

56.

Lahiri

(2010). Geographic distribution of R&D activity: How does it affect innovation quality? Academy of Management Journal, 53(5), 1194–1209. https://doi.org/10.5465/amj.2010.54533233

57.

Lai

D’Amour

Fleming

(2011). Disambiguation and co-authorship networks of the U.S. Patent inventor database (1975-2010). https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/5F1RRI

58.

Lanjouw

J. O.

Schankerman

(2004). Patent quality and research productivity: Measuring innovation with multiple indicators. The Economic Journal, 114(495), 441–465. https://doi.org/10.1111/j.1468-0297.2004.00216.x

59.

Lau

D. C.

Murnighan

J. K.

(1998). Demographic diversity and faultlines: The compositional dynamics of organizational groups. Academy of Management Review, 23(2), 325–340. https://doi.org/10.5465/amj.2005.17843943

60.

Leahey

Beckman

C. M.

Stanko

T. L.

(2017). Prominent but less productive: The impact of interdisciplinarity on scientists’ research. Administrative Science Quarterly, 62(1), 105–139. https://doi.org/10.1177/0001839216665364

61.

Lee

Y. N.

Walsh

J. P.

Wang

(2015). Creativity in scientific teams: Unpacking novelty and impact. Research Policy, 44(3), 684–697. https://doi.org/10.1016/j.respol.2014.10.007

62.

Lehmann-Willenbrock

Hung

Keyton

(2017). New frontiers in analyzing dynamic group interactions: Bridging social and computer science. Small Group Research, 48(5), 519–531. https://doi.org/10.1177/1046496417718941

63.

Levinthal

D. A.

March

J. G

. (1993). The myopia of learning. Strategic Management Journal, 14(Suppl. 2), 95–112. https://doi.org/10.1002/smj.4250141009

64.

Lewis

(2003). Measuring transactive memory systems in the field: Scale development and validation. Journal of Applied Psychology, 88(4), 587–604. https://doi.org/10.1037/0021-9010.88.4.587

65.

Magee

J. C.

Galinsky

A. D.

(2008). Social hierarchy: The self-reinforcing nature of power and status. The Academy of Management Annals, 2(1), 351–398. https://doi.org/10.5465/19416520802211628

66.

Mathieu

J. E.

Heffner

T. S.

Goodwin

G. F.

Salas

Cannon-Bowers

(2000). The influence of shared mental models on team process and performance. Journal of Applied Psychology, 85(2), 273–283. https://dx-doi-org.web.bisu.edu.cn/10.1037/0021-9010.85.2.273

67.

Mello

A. L.

Rentsch

J. R.

(2015). Cognitive diversity in teams: A multidisciplinary review. Small Group Research, 46(6), 623–658. https://doi.org/10.1177/1046496415602558

68.

Mesmer-Magnus

J. R.

DeChurch

L. A.

(2009). Information sharing and team performance: A meta-analysis. Journal of Applied Psychology, 94(2), 535–546. https://doi.org/10.1037/a0013773

69.

Mesmer-Magnus

J. R.

DeChurch

L. A.

Jimenez-Rodriguez

Wildman

Shuffler

(2011). A meta-analytic investigation of virtuality and information sharing in teams. Organizational Behavior and Human Decision Processes, 115(2), 214–225. https://doi.org/10.1016/j.obhdp.2011.03.002

70.

Mohammed

Klimoski

Rentsch

J. R.

(2000). The measurement of team mental models: We have no shared schema. Organizational Research Methods, 3(2), 123–165. https://doi.org/10.1177/109442810032001

71.

Nembhard

I. M.

Edmondson

A. C.

(2006). Making it safe: The effects of leader inclusiveness and professional status on psychological safety and improvement efforts in health care teams. Journal of Organizational Behavior, 27(7), 941–966. https://doi.org/10.1002/job.413

72.

Peters

L. S.

(2015). The continuing economic potential of nanotechnology. In Wiley Encyclopedia of management (pp. 1–18). https://doi.org/10.1002/9781118785317.weom130060

73.

Polzer

J. T.

Crisp

C. B.

Jarvenpaa

S. L.

Kim

J. W.

(2006). Extending the faultline model to geographically dispersed teams: How colocated subgroups can impair group functioning. Academy of Management Journal, 49(4), 679–692. https://doi.org/10.5465/amj.2006.22083024

74.

Reagans

Argote

Brooks

(2005). Individual experience and experience working together: Predicting learning rates from knowing who knows what and knowing how to work together. Management Science, 51(6), 869–881. https://doi.org/10.1287/mnsc.1050.0366

75.

Rico

Molleman

Sánchez-Manzanares

Van der Vegt

G. S.

(2007). The effects of diversity faultlines and team task autonomy on decision quality and social integration. Journal of Management, 33(1), 111–132. https://doi.org/10.1177/0149206306295307

76.

Rico

Sánchez-Manzanares

Gil

Gibson

(2008). Team implicit coordination processes: A team knowledge–based approach. Academy of Management Review, 33(1), 163–184. https://doi.org/10.5465/amr.2008.27751276

77.

Riebe

Girardi

Whitsed

(2016). A systematic literature review of teamwork pedagogy in higher education. Small Group Research, 47(6), 619–664. https://doi.org/10.1177/1046496416665221

78.

Rosenkopf

Almeida

(2003). Overcoming local search through alliances and mobility. Management Science, 49(6), 751–766. https://doi.org/10.1287/mnsc.49.6.751.16026

79.

Salazar

M. R.

Lant

T. K.

Fiore

S. M.

Salas

(2012). Facilitating innovation in diverse science teams through integrative capacity. Small Group Research, 43(5), 527–558. https://doi.org/10.1177/1046496412453622

80.

Schriesheim

C. A.

Castro

S. L.

Cogliser

C. C.

(1999). Leader-member exchange (LMX) research: A comprehensive review of theory, measurement, and data-analytic practices. The Leadership Quarterly, 10(1), 63–113. https://doi.org/10.1016/S1048-9843(99)80009-5

81.

Sorensen

Stuart

T. E.

(2000). Aging, obsolescence, and organizational innovation. Administrative Science Quarterly, 45, 81–112. https://doi.org/10.2307/2666980

82.

Srivastava

M. K.

Gnyawali

D. R.

(2011). When do relational resources matter? Leveraging portfolio technological resources for breakthrough innovation. Academy of Management Journal, 54(4), 797–810. https://doi.org/10.5465/amj.2011.64870140

83.

Stasser

Titus

(1985). Pooling of unshared information in group decision making: Biased information sampling during discussion. Journal of Personality and Social Psychology, 48(6), 1467–1478. https://doi.org/10.1037/0022-3514.48.6.1467

84.

Stasser

Vaughan

S. I.

Stewart

D. D.

(2000). Pooling unshared information: The benefits of knowing how access to information is distributed among group members. Organizational Behavior and Human Decision Processes, 82(1), 102–116. https://doi.org/10.1006/obhd.2000.2890

85.

Taylor

Greve

H. R.

(2006). Superman or the fantastic four? Knowledge combination and experience in innovative teams. Academy of Management Journal, 49(4), 723–740. https://doi.org/10.5465/amj.2006.22083029

86.

Trajtenberg

(1990). A penny for your quotes: Patent citations and the value of innovations. The Rand Journal of Economics, 21(1), 172–187. https://www.jstor.org/stable/2555502

87.

Tzabbar

(2009). When does scientist recruitment affect technological repositioning? Academy of Management Journal, 52(5), 873–896. https://doi.org/10.5465/amj.2009.44632853

88.

Tzabbar

Vestal

(2015). Bridging the social chasm in geographically distributed R&D teams: The moderating effects of relational strength and status asymmetry on the novelty of team innovation. Organization Science, 26(3), 811–829. https://doi.org/10.1287/orsc.2015.0969

89.

Uzzi

B. D.

(1997). Social structure and competition in interfirm networks: The paradox of embeddedness. Administrative Science Quarterly, 42(1), 35–67. http://doi.org/10.2307/2393808

90.

Uzzi

B. D.

Lancaster

(2003). Relational embeddedness and learning: The case of bank loan managers and their clients. Management Science, 49(4), 383–399. https://doi.org/10.1287/mnsc.49.4.383.14427van

91.

der Vegt

G. S.

De Jong

S. B.

Bunderson

J. S.

Molleman

(2010). Power asymmetry and learning in teams: The moderating role of performance feedback. Organization Science, 21(2), 347–361. https://doi.org/10.1287/orsc.1090.0452

92.

van Dijk

van Engen

M. L.

van Knippenberg

. (2012). Defying conventional wisdom: A meta-analytical examination of the differences between demographic and job-related diversity relationships with performance. Organizational Behavior and Human Decision Processes, 119(1), 38–53. https://doi.org/10.1016/j.obhdp.2012.06.003

93.

van Knippenberg

. (2017). Team innovation. Annual Review of Organizational Psychology and Organizational Behavior, 4, 211–233. https://doi.org/10.1146/annurev-orgpsych-032516-113240

94.

van Knippenberg

Dawson

J. F.

West

M. A.

Homan

A. C

. (2011). Diversity faultlines, shared objectives, and top management team performance. Human Relations, 64(3), 307–336. https://doi.org/10.1177/0018726710378384

95.

van Knippenberg

De Dreu

C. K.

Homan

A. C

. (2004). Work group diversity and group performance: An integrative model and research agenda. Journal of Applied Psychology, 89(6), 1008–1022. https://doi.org/10.1037/0021-9010.89.6.1008

96.

Vera

Crossan

(2005). Improvisation and innovative performance in teams. Organization Science, 16(3), 203–224. https://doi.org/10.1287/orsc.1050.0126

97.

Wernerfelt

(1984). A resource-based view of the firm. Strategic Management Journal, 5(2), 171–180. https://doi.org/10.1002/smj.4250050207

98.

West

M. A.

(2002). Sparkling fountains or stagnant ponds: An integrative model of creativity and innovation implementation in work groups. Applied Psychology, 51(3), 355–387. https://doi.org/10.1111/1464-0597.00951

99.

Wiersema

M. F.

Bowen

H. P.

(2009). The use of limited dependent variable techniques in strategy research: Issues and methods. Strategic Management Journal, 30(6), 679–692. https://doi.org/10.1002/smj.758

100.

Xiao

Zhang

Basadur

T. M.

(2016). Does information sharing always improve team decision making? An examination of the hidden profile condition in new product development. Journal of Business Research, 69(2), 587–595. https://doi.org/10.1016/j.jbusres.2015.05.014

101.

Xie

X. Y.

Wang

W. L.

Luan

(2014). It is not what we have, but how we use it: Reexploring the relationship between task conflict and team innovation from the resource-based view. Group Processes & Intergroup Relations, 17(2), 240–251. https://doi.org/10.1177/1368430213502559

102.

Yong

Sauer

S. J.

Mannix

E. A.

(2014). Conflict and creativity in interdisciplinary teams. Small Group Research, 45(3), 266–289. https://doi.org/10.1177/1046496414530789

103.

Zelner

B. A.

(2009). Using simulation to interpret results from logit, probit, and other nonlinear models. Strategic Management Journal, 30(12), 1335–1348. https://doi.org/10.1002/smj.783

104.

Zhang

(2006). The dynamics of online collaboration: Team task, team development, peer relationship, and communication media. In Afonso

A. P.

Figueiredo

A. D.

(Eds.), Managing learning in virtual settings: The role of context (pp. 98–116). IGI Global.

105.

Zhang

Z. X.

Hempel

P. S.

Han

Y. L.

Tjosvold

(2007). Transactive memory system links, work team characteristics, and performance. Journal of Applied Psychology, 92(6), 1722–1730. https://doi.org/10.1037/0021-9010.92.6.1722

106.

Zucker

L. G.

Darby

M. R.

(2009). Nanobank data description, release 2.0 (beta-test). UCLA Center for International Science, Technology, and Cultural Policy.

107.

Zucker

L. G.

Darby

M. R.

(2011). COMETS data description, release 1.0. UCLA Center for International Science, Technology, and Cultural Policy.