High self-monitors—who aim to make a good impression and fit in, chameleon-like, in different situations rather than stay consistent across situations—tend to occupy broker positions in the network structure. (Sasovova, Mehra, Borgatti, & Schippers, 2010, as cited by Kwon, Rondi, Levin, De Massis, and Brass (2020))
Introduction
This proposal aims to investigate the effects of status and gender as the antecedents of brokerage formation. Brokerage has been boomingly discussed by management scholars, yet the majority of the literature focuses on brokerage structure or outcome (as reviewed by Kwon et al., 2020). For instance, Burt (1992, as cited by Zaheer and Soda (2009)) argues that a network position rich in structural holes, by which an actor is connected to a large number of disconnected alters, might be beneficial in establishing a positive reputation. On the contrary, brokerage antecedent was discussed only recently by few scholars (Boari & Riboldazzi, 2014; Kleinbaum, Jordan, & Audia, 2015; Zaheer & Soda, 2009). Boari and Riboldazzi (2014) follow Lee (2010) to postulate a positive causal relationship between status and brokerage formation. However, the linear relation they argue is against my personal experience. Although node of higher status (e.g., manager) is more likely to occupy brokerage, node whose status is lower than its alters also have good chance to become a broker. For example, executive managers are often brokered by an office assistant. Hence, we postulate a contingent effect of status on brokerage formation. In addition, gender has been largely ignored in the studies of network formation, argues Whittington (2018) who was amongst the first scholars in investigating gender’s effect on network formation. Therefore, we include gender as a moderator to this contingent relationship.
The empirical context of this proposal is academic collaboration that has several advantages. First, it is relevant to the information benefit of brokerage since academic performance is built upon knowledge (e.g., novelty of ideas and information flowing from nonredundant ties (Burt, 2004, as cited in Lee (2010)). Second, publishing relationships have been scantly studied as most published research focuses on commercial collaboration network such as patent (Whittington, 2018). Third, academic collaboration has a unique strength in brokerage antecedent research because node attributes (e.g., gender; and status that is proxied by university ranking) are ex ante at the time of tie formation (i.e., publication co-authorship), providing a natural time lag. Lastly, this context is methodologically feasible, allowing researcher to circumvent the common obstacle of data access in interpersonal network research because most of authors’ information is readily available on the internet.
The rest of this proposal is organized as follow: in theory development section we discuss the inconsistency of brokerage antecedents research along with my hypotheses, in the preliminary analysis and results section we report results from an empirical application, and in the discussion we offer directions for future works.
Theory development and Hypotheses
Previous studies investigate career history (Kleinbaum 2012), training (Burt and Ronchi 2007), and organizational structure (Kleinbaum and Stuart 2014) as antecedents of brokerage (as reivewed by Kleinbaum et al., 2015). However, as Stuart and Sorenson (2007) point out the endogeneity issue of network formation: “Extant research has almost entirely ignored this rampant endogeneity problem and, as a result, bias likely contaminates many (if not most) of the existing estimates of network effects”, such issue also happens to brokerage antecedent research. For example, Powell et al. (2005: 1140, as cited by Zaheer and Soda (2009)) argue that central and high-status actors are likely to receive a disproportionate share of future ties. What makes a node to become a broker is, therefore, a valid research question (Kwon et al., 2020).
Status has often been discussed with brokerage despite inconsistent findings (Brass, Butterfield, & Skaggs, 1998). On one hand, as Podolny (1993) uses Bonacich’s (1987) two-parameter centrality measure to proxy for actors’ statuses (Stuart & Sorenson, 2007), actor with structural holes is, by definition, prominent in the overall network (Boari & Riboldazzi, 2014). Both Zuckerman (1967); Zucker and Darby (2006) find that high-quality scientists are more likely to form collaborative relationships (Lee, 2010, whose findings are similar using firm-level collaboration in biotechnology). Likewise, Sauder, Lynn, & Podolny (2012, as reviewed by Kwon et al. (2020)) suggest that status leads to brokerage formation. On the other hand, Podolny (2005: 233, as citd by Zaheer and Soda (2009)) suggests that reaching out to lower-status players may increase structural holes but reduce status, implying a negative relationship between status and structural holes. Similarly, Chandler, Haunschild, Rhee, and Beckman (2013) find a negative link between status and brokerage in firm-level. In an attempt to reconcile this inconsistency, we present the Hypothesis 1 as follow:
Hypothesis 1: The relationship between status and brokerage formation is U-shaped
Status structure is often related to hierarchical gender relations (Fox et al., 2017, as cited by Whittington (2018)). Amongst the first scholars who adds gender into the interplay of status and network formation, Whittington (2018) points out that “men hold almost twice the number of brokerage positions than women do across their years in the network” based on evidence from science inventor collaboration. Moreover, gender is also related to status, where women have greater status-asymmetries between themselves and their co-inventors (Whittington, 2018). Hence, we present the Hypothesis 2 as follow:
Hypothesis 2: Gender moderates the relation between status and brokerage formation
Reference
Boari, C., & Riboldazzi, F. 2014. How knowledge brokers emerge and evolve: The role of actors’ behaviour. Research Policy, 43(4): 683-695.
Bothner, M. S., Kim, Y.-K., & Smith, E. B. 2012. How does status affect performance? Status as an asset vs. status as a liability in the PGA and NASCAR. Organization Science, 23(2): 416-433.
Brass, D. J., Butterfield, K. D., & Skaggs, B. C. 1998. Relationships and unethical behavior: A social network perspective. Academy of management review, 23(1): 14-31.
Haans, R. F., Pieters, C., & He, Z. L. 2016. Thinking about U: Theorizing and testing U‐and inverted U‐shaped relationships in strategy research. Strategic Management Journal, 37(7): 1177-1195.
Halevy, N., Halali, E., & Zlatev, J. J. 2019. Brokerage and brokering: An integrative review and organizing framework for third party influence. Academy of Management Annals, 13(1): 215-239.
Kleinbaum, A. M., Jordan, A. H., & Audia, P. G. 2015. An altercentric perspective on the origins of brokerage in social networks: How perceived empathy moderates the self-monitoring effect. Organization Science, 26(4): 1226-1242.
Kwon, S.-W., Rondi, E., Levin, D. Z., De Massis, A., & Brass, D. J. 2020. Network brokerage: An integrative review and future research agenda. Journal of Management, 46(6): 1092-1120.
Lee, J. J. 2010. Heterogeneity, brokerage, and innovative performance: Endogenous formation of collaborative inventor networks. Organization Science, 21(4): 804-822.
Smith, E. B., Menon, T., & Thompson, L. 2012. Status differences in the cognitive activation of social networks. Organization Science, 23(1): 67-82.
Stern, I., Dukerich, J. M., & Zajac, E. 2014. Unmixed signals: How reputation and status affect alliance formation. Strategic Management Journal, 35(4): 512-531.
Stuart, T. E., & Sorenson, O. 2007. Strategic networks and entrepreneurial ventures. Strategic Entrepreneurship Journal, 1(3‐4): 211-227.
Whittington, K. B. 2018. A tie is a tie? Gender and network positioning in life science inventor collaboration. Research Policy, 47(2): 511-526.
Zaheer, A., & Soda, G. 2009. Network evolution: The origins of structural holes. Administrative Science Quarterly, 54(1): 1-31.
Acknowledgement
The author has collected the dataset under the research assistantship funding from Professor Jin-Su Kang, who grants permission to use the dataset for this proposal.