running head: hierarchy and team effectiveness meta-analysisset each other. second, we integratively...

Running head: Hierarchy and Team Effectiveness Meta-Analysis

Why and When Hierarchy Impacts Team Effectiveness:

A Meta-Analytic Integration

Lindred L. Greer*, Bart A. de Jong, Maartje E. Schouten, and Jennifer Dannals

Journal of Applied Psychology, in press

* To whom correspondence should be addressed. Address: Stanford Graduate School of Business, Knight Management Center, 655 Knight Way, Stanford, CA, 94305. E-mail: [email protected]. **We would like to think our reviewers as well as Drew Carton, Katy DeCelles, Frank Flynn, Nir Halevy, John Hollenbeck, Russ Johnson, Lara Tiedens, and Lisanne Van Bunderen for their constructive comments in developing this article. ***Financial support was granted to the first author by the Netherlands Science Foundation (451-11-034) and to the second author by the Netherlands Science Foundation (016-145-243).

©American Psychological Association, 2017. This paper is not the copy of record and may not exactly replicate the authoritative document published in the APA journal. Please do not copy or cite without author's permission. The final article is available, upon publication, at (awaiting DOI).

Hierarchy and Team Effectiveness Meta-Analysis 2

Abstract

Hierarchy has the potential to both benefit and harm team effectiveness. In this paper, we meta-

analytically investigate different explanations for why and when hierarchy helps or hurts team

effectiveness, drawing on results from 54 prior studies (N = 13,914 teams). Our findings show

that, on net, hierarchy negatively impacts team effectiveness (performance: r = -.08; viability: r

= -.11), and that this effect is mediated by increased conflict-enabling states. Additionally, we

show that the negative relationship between hierarchy and team performance is exacerbated by

aspects of the team structure (i.e., membership instability, skill differentiation) and the hierarchy

itself (i.e., mutability), which make hierarchical teams prone to conflict. The predictions

regarding the positive effect of hierarchy on team performance as mediated by coordination-

enabling processes, and the moderating roles of several aspects of team tasks (i.e.,

interdependence, complexity) and the hierarchy (i.e., form) were not supported, with the

exception that task ambiguity enhanced the positive effects of hierarchy. Given that our findings

largely support dysfunctional views on hierarchy, future research is needed to understand when

and why hierarchy may be more likely to live up to its purported functional benefits.

Keywords: Hierarchy, Teams, Meta-Analysis, Power, Status, Conflict


Why and When Hierarchy Impacts Team Effectiveness:

A Meta-Analytic Integration

Hierarchy, defined as vertical differences between members in their possession of socially

valued resources (Harrison & Klein, 2007; Hays & Bendersky, 2015), is a fundamental concept

in the study of groups and teams (for reviews, see Anderson & Brown, 2010; Halevy, Chou, &

Galinsky, 2011; Keltner, Van Kleef, Chen, & Kraus, 2008; Magee & Galinsky, 2010). Past

theory and research offers divergent perspectives on how and why hierarchy impacts team

effectiveness. The functionalist perspective proposes a positive pathway from hierarchy to team

effectiveness via improved coordination-enabling processes (e.g., Halevy et al., 2011; Keltner et

al., 2008; Van der Heijden, Potters, & Sefton, 2009), while the conflict perspective proposes a

negative pathway via increased conflict-enabling states (e.g., Bloom, 1999; Greer & Van Kleef,

2010; Greer, Van Bunderen, & Yu, 2017; Wolfe & McGinn, 2005). To understand these

potentially off-setting processes and identify when one or the other is more likely to take priority,

a growing line of work has started to identify different contingency factors that explain when

hierarchy is most likely to help or hurt team effectiveness (e.g., Anderson & Willer, 2014;

Bunderson, Van der Vegt, Cantimur, & Rink, 2016; Hays & Bendersky, 2015; Tarakci, Greer, &

Groenen, 2016). These contingency factors include task characteristics (e.g., interdependence;

Ronay, Greenaway, Anicich, & Galinsky, 2012), team structure (e.g., skill differentiation;

Young-Hyman, 2017), and team hierarchy (e.g., hierarchy form; Bunderson et al., 2016).

While scholarly work in this area is burgeoning, the field currently lacks a cumulative

investigation of the different, and potentially off-setting, explanations for why hierarchy impacts

team effectiveness. Studies have thus far progressed in a disconnected and piecemeal fashion,

with studies focusing on one pathway at a time, to the exclusion of others (e.g., Bunderson et al.,


2016). This is problematic, as singling out a specific mechanism risks overstating the importance

of that mechanism (De Jong & Elfring, 2010), prevents scholars from gauging the relative

adequacy of different mediating mechanisms (Colquitt et al., 2013), and does not allow scholars

to examine the extent to which these mechanisms complement or off-set each other. Similarly,

the field lacks a cumulative investigation of contingency factors explaining the conditions under

which hierarchy helps or hurts team effectiveness, as research has also only tended to examine a

single contingency factor at a time (e.g., Ronay et al., 2012). Given the multitude of contingency

factors proposed so far, this lack of integrative testing may contribute to the proliferation of

unnecessary theoretical complexity (Banks, McCauley, Gardner, & Guler, 2016), by preventing

scholars from gauging the relative predictive power of any specific moderator (Ng, 2017).

Therefore, while the rapidly expanding work in this area is encouraging, the time has come to

integrate and empirically compare the multiple proposed mediating mechanisms as well as

integrate and empirically compare the multiple contingency factors (Bergh et al., 2014; Chan &

Arvey, 2012).

The aim of this study is two-fold. First, we seek to integrate the two key pathways

explaining why hierarchy has been proposed to impact team effectiveness into a single

theoretical and empirical model to better understand how these pathways may co-exist and off-

set each other. Second, we integratively examine and compare a comprehensive set of

contingency factors proposed by scholars to explain when the positive versus negative effects of

hierarchy on team effectiveness will be more likely to take priority (e.g., Anderson & Brown,

2010; Anderson & Willer, 2014; Bunderson et al., 2016; Hays & Bendersky, 2015; Halevy et al.,

2011). We consider these issues of why and when independently, and empirically test each of

them via a meta-analysis, which integrates findings from 54 primary studies on hierarchy and


team effectiveness (involving 86 effect sizes, and 13,914 teams) conducted over the past six

decades. In doing so, we aspire to provide the field with improved clarity, consistency, and

predictive ability in construct and theory development, and help the field move away from theory

proliferation absent explanatory power (Banks et al., 2016; Leavitt, Mitchell, & Peterson, 2010;

LePine & Wilcox-King, 2010; Schlaegel & Koenig, 2014). Such integration can help provide

insight into which perspectives on hierarchy may need qualification and give suggestions for

how other perspectives may need changed or qualified.

Why Hierarchy Impacts Team Effectiveness

We seek here to identify and integrate the key explanations for why, as well as for when,

hierarchy impacts team effectiveness. Consistent with the teams literature (Hackman, 1987;

Kozlowski & Ilgen, 2006), team effectiveness is represented in this paper by two key indicators

– team performance, or the degree to which a team accomplishes its goals or mission, as

reflected by task-indicators such as team output quality, quantity, and efficiency (Bell, 2007;

Devine & Phillips, 2001), and team viability, or the ability of the team to continue in the future

based on member willingness to remain with the team, as reflected by socio-affective indicators,

such as team satisfaction, commitment, and turnover intentions (Balkundi & Harrison, 2006).

In explaining the hierarchy-effectiveness relationship, the functionalist perspective

proposes a positive pathway between hierarchy and team effectiveness via improved

coordination-enabling processes (e.g., Anderson, Srivastava, Beer, Spataro, & Chatman, 2006;

Keltner et al., 2008; Van Vugt et al., 2008), which are defined as the behaviors and strategies

used to integrate and align individual member actions, knowledge, and objectives towards the

attainment of common goals (Arrow, McGrath, & Berdahl, 2000; Rico, Sanchez-Manaznares,

Gil, & Gibson, 2008; Zalesny, Salas, & Prince, 1995). The functionalist line of thought stems


from research in social psychology, which suggests that people have an unconscious preference

for hierarchy (Tiedens, & Fragale, 2003) because of the comfort it gives in coordinating team

interactions (Halevy et al., 2011). Scholars in this line of thought argue that hierarchy may

benefit team effectiveness by facilitating and coordinating member interactions (e.g., De

Kwaadsteniet, Van Dijk, Wit, De Cremer, & de Rooij, 2007; Galinsky, Chou, Halevy, & Van

Kleef, 2012; Woolley, Gerbassi, Chabris, Kosslyn, & Hackman, 2008). This can include guiding

resource allocation (Keltner et al., 2008), allowing members a means of knowing their place

within the team (Anderson & Brown, 2010), and clarifying expectations about norms, roles, and

expected behaviors based on members’ placement in the hierarchy (e.g., Biggart & Hamilton.

1984; Dornbusch & Scott, 1975). By providing a structure to guide roles and interactions,

hierarchy may enhance coordination-enabling processes and thereby improve team effectiveness.

For example, Woolley and her colleagues (2008) found in a laboratory experiment of student

teams, that when teams had a clear expertise hierarchy, they were better able to integrate

information and solve a hypothetical terrorist plot. Such coordination-enabling processes in turn

benefit team effectiveness by facilitating expedient and effective combining of individual inputs

to a team outcome (e.g., Cohen & Bailey, 1997; Gladstein, 1984). Indeed, meta-analytic

evidence has shown that coordination-enabling processes have a positive impact on both team

performance and viability (LePine, Piccolo, Jackson, Mathieu, & Saul, 2008). Taken together,

these theoretical arguments and empirical evidence suggest that hierarchy can offer teams a

positive pathway to team effectiveness through improved coordination-enabling processes.

In contrast, the conflict perspective proposes a negative pathway from hierarchy to team

effectiveness via heightened conflict-enabling states (e.g., Bunderson & Reagans, 2011;

DeBrock, Hendricks, & Koenker, 2004; Greer et al., 2017; Taracki et al., 2016), which are


defined as emergent team states characterized by the potential for perceived incompatibilities or

differences between members (De Dreu & Gelfand, 2008; De Wit et al., 2012; Jehn, 1995).

According to this perspective, hierarchy may motivate individuals to climb up the ranks (Magee

& Galinsky, 2008), and may lead members at different ranks to have opposing interests and

perspectives (Van Bunderen, Greer, & Van Knippenberg, 2017), which can lead to increased

intragroup conflicts (over rank: Anderson & Brown, 2010; Greer & Dannals, 2017; Hays &

Bendersky, 2015; and generally: Bunderson et al., 2016). Such conflicts have been shown to

harm team outcomes (e.g., conflicts over rank: Bendersky & Hays, 2012; Ronay et al., 2012;

conflicts more generally: for meta-analyses, see De Dreu & Weingart, 2003; De Wit et al., 2012),

distracting members from task accomplishment and harming interpersonal relationships. In

support of this perspective, for example, Greer and Van Kleef (2010) found that hierarchy

harmed team performance in high-power teams because hierarchy provoked contests and

conflicts over the rank order in the team. Indeed, a number of empirical studies have indeed now

shown that hierarchy can create a tense intra-team environment, in which teams have high levels

of conflict states and lowered team outcomes (e.g., Edmondson, 2002; Mannix, 1993; Torrance,

1955; Tost, Gino, & Larrick, 2013; Van der Vegt, De Jong, Bunderson, & Molleman, 2010;

Wolfe & McGinn, 2005). Taken together, this line of work suggests that hierarchy can harm

team effectiveness via conflict-enabling states.

Given the co-existence of these two off-setting pathways in the literature, we propose:

Hypothesis 1: Coordination-enabling processes mediate a positive relationship between

hierarchy and (a) team performance and (b) team viability, even after controlling for the

mediating effect of conflict-enabling states.

Hypothesis 2: Conflict-enabling states mediate a negative relationship between hierarchy


and (a) team performance and (b) team viability, even after controlling for the mediating

effect of coordination-enabling processes.

When Hierarchy Impacts Team Performance

Recognizing the need to reconcile the off-setting processes which have been proposed to

explain why hierarchy affects team effectiveness, and to understand when one process will take

priority over the other, scholars have argued for the need to adopt a ‘bounded account of

hierarchy’ (e.g., Anderson & Willer, 2014; Bunderson et al., 2016), and have begun to identify

contingency factors that could explain the conditions under which hierarchy is beneficial or

detrimental for teams. For example, Halevy, Chou, Galinsky, and Murnighan (2012) suggest that

hierarchy benefits performance in teams with high coordination demands (i.e., high task

interdependence), and Bunderson et al. (2016) show that the conflictual downsides of hierarchy

are worse when the hierarchical form is based more on inequity or centralization than acyclical

patterns of dyadic influence. Given that the majority of contingency findings have been proposed

for the relationship between hierarchy and performance (and not viability), we focus our

theorizing here on team performance, but encourage future research to examine contingencies of

the hierarchy–team viability relationship as well.

In integrating specific contingency factors, we purposefully build on pre-existing

contingency frameworks that have been proposed in the literature (e.g., Anderson & Brown,

2010; Bunderson et al., 2016; Greer et al., 2017; Hays & Bendersky, 2015; Halevy et al., 2011).

To ensure that our contingency factors are comprehensive and meaningful as a set, we focus on

categories of factors that are conceptually distinct in nature and functionally distinct in terms of

which pathway between hierarchy and team performance they cause to dominate. With regard to

conceptual distinctions, we draw on past team meta-analyses and differentiate between


contingency factors that pertain to aspects of team tasks (e.g., task interdependence, LePine et al,

2008), aspects of team structure (e.g., skill differentiation, De Jong, Dirks, & Gillespie, 2016),

and aspects of the focal independent variable itself (e.g., form of diversity, Bell et al., 2011), in

our case hierarchy. With regard to functional distinctions, we link our contingency factors to the

two off-setting pathways identified in the literature and differentiate between factors that are

predicted to give priority to the positive pathway between hierarchy and performance by

increasing coordination demands (e.g., task complexity, Bunderson et al., 2016), and factors that

are predicted to give priority to the negative pathway by increasing conflict susceptibilities (e.g.,

hierarchy mutability, Hays & Bendersky, 2015). Below, we describe each specific contingency

factor, and our predictions for how it will affect the relationship between hierarchy and team

performance.

Team Task Contingencies

Theory and research on hierarchy has led to the expectation that hierarchy will most

benefit performance when teams work on tasks with high coordination demands (Anderson &

Brown, 2010; Halevy et al., 2011). Not all teams require coordination to perform well, but for

teams where aspects of the team’s task place high demands on members to integrate and align

individual member actions, knowledge, and objectives towards the attainment of common goals

(Arrow et al., 2000; Rico et al., 2008), we expect hierarchy to be the most useful. Hierarchy can

help meet coordination demands by providing a heuristic to allocate resources in the team

(Keltner et al., 2008), to integrate individual efforts as members will know who to defer to for

what (Halevy et al., 2011), and to process information and provide a structure to organize

information flow, ensuring no valuable knowledge gets lost. When teams in high need of

coordination are able to coordinate individual contributions well, they are likely to perform better


(Steiner, 1972). Thus, as implied by the functionalist perspective, hierarchy should benefit teams

when coordination demands are high. In contrast, in teams with low coordination demands,

hierarchy may become superfluous and unlikely to positively impact team performance.

We focus here on three aspects of team tasks which increase the utility of such

coordination-enabling processes and have been commonly investigated in the hierarchy

literature: task interdependence, task ambiguity, and task complexity. Task interdependence is

defined as the degree to which members need to rely on and connect with one another for task

accomplishment (Shea & Guzzo, 1987; Saavedra, Earley, & Van Dyne, 1993). Task

interdependence increases coordination demands on teams (Saavedra et al., 1993), as when

members are closely bound together to accomplish team objectives, they have to communicate

more frequently, make common decisions on team task components, and sync individual task

efforts with each other. In such situations, the structure and mental-map provided by hierarchy

can help coordinate these interactions, enabling higher team performance. Scholars in this area

support this reasoning, as they have theorized and found that hierarchy is more positively related

to performance when interdependence is high (e.g., Halevy et al., 2012; Ronay et al., 2012).

Task ambiguity is defined as the degree to which members lack clarity about their

responsibilities and what is expected of them to accomplish a task (Kahn, Wolfe, Quinn, Snoek,

& Rosenthal, 1964). Task ambiguity can increase the coordination demands on the team

(Barton, 1980; Valentine & Edmondson, 2014). When task ambiguity is high, team members

have high levels of uncertainty about how they should exert their work-related efforts, and how

their efforts should align with others. In such situations, hierarchy can provide a heuristic to help

guide and structure member activities and interactions. Given its relevance to the study of

hierarchy and structure, task ambiguity has indeed been proposed as a moderator of the effects of


hierarchy on team performance (e.g., Pandey & Rainey, 2006).

Task complexity implies the presence of multiple potential pathways to achieve a goal,

the presence of multiple desired outcomes, the presence of conflicting interdependence among

the different paths to achieve goals, and the presence of probabilistic links between paths and

outcomes (Campbell, 1988). Task complexity has been shown to increase the coordination

demands on teams (e.g., Zhou, 2013). When task complexity is high, team members have a high

information load, and anything which can simplify their responsibilities and guide their

interactions can be useful to facilitate task accomplishment. As such, task complexity has indeed

been shown to moderate the effects of hierarchy on performance (e.g., Bunderson et al., 2016).

Consistent with this theorizing and empirical evidence, we hypothesize:

Hypothesis 3: Task interdependence (a), task ambiguity (b), and task complexity (c)

moderate the positive relationship between hierarchy and performance, such that the

relationship becomes stronger as the levels of these task characteristics increase.

Team Structure Contingencies

In understanding how team structure may affect the relationship between hierarchy and

team performance, we build on emerging trends in the hierarchy literature which suggest that

hierarchy’s downsides are most likely when conflicts emerge in the team (Bunderson et al.,

2016; Greer & Van Kleef, 2010). While not all teams are particularly prone to conflict, in teams

whose structures leave them particularly vulnerable to conflict, we expect to see the most

negative effect of hierarchy on team performance. The reasoning in this work derives from the

conflict perspective on hierarchy (Tarakci et al., 2016), which suggests that in teams whose

structures are susceptible to conflicts, the rank differences created by hierarchy may be

especially likely to trigger intense conflicts and team performance is likely to suffer. Namely,


when teams are already susceptible to conflict, hierarchies may become a bone of contention to

activate underlying tensions, such as between different functional subgroups in highly skill-

differentiated teams. In such situations, differences in ranks and resources may be particularly

salient and resented, as they will be paired with underlying categorical in-group/out-group

distinctions, which can cause members to engage in conflicts over ranks and resources. Such

hierarchical conflicts have been shown to distract members, impair conflict resolution process,

and harm team performance (Bendersky & Hays, 2012; Chun & Choi, 2014; Greer & Van Kleef,

2010). Therefore, teams with structures more susceptible to internal conflict-enabling states are

more likely to have negative effects of hierarchy on team performance.

Besides hierarchy itself (what Hollenbeck, Beersma, & Schouten [2012] term ‘authority

differentiation’), we focus on two other team structural characteristics that together comprise the

big-three team structural characteristics, namely membership instability and skill differentiation.

Membership instability is defined as teams with frequent membership changes (Hollenbeck et

al., 2012). When teams have frequent membership changes, and members enter or leave or the

team as a whole forms and dissolves quickly, the hierarchy has to be frequently renegotiated.

This ongoing conflict can lead to rivalries and tensions among different members, and the

presence of a hierarchy in such a volatile situation can explode existing tensions and fuel future

rivalries. Such frequently changing teams may instead benefit from flatter structures, where less

negotiation is required as members enter or leave the team. Membership instability has been

shown to increase a team’s conflict susceptibilities (e.g., Tuckman, 1965), and has been

suggested to have implications for the utility of hierarchical team structures (e.g., Valentine &

Edmondson, 2014).

Skill differentiation is defined as the degree to which members have specialized


knowledge or functional capacities that make it difficult for members to be substituted and this is

reflected in differences in members’ educational major, area of functional work experience,

and/or expertise area (Hollenbeck et al., 2012). Skill differentiation has been shown to increase a

team’s conflict susceptibilities (Bunderson & Sutcliffe, 2002; Cronin & Weingart, 2007;

Dougherty, 1992). When teams consist of members with different skill sets, underlying tensions

can exist between different functional groups (Dougherty, 1992). Hierarchy can then trigger such

tensions by creating inequities and resentments between these different subgroups. Skill

differentiation has indeed been suggested in the literature on hierarchy to have implications for

the utility and effectiveness of hierarchical structures (e.g., Gruenfeld & Tiedens, 2010; Young-

Hyman, 2017). Based on these lines of theorizing and empirical findings, we predict:

Hypothesis 4: Membership instability (a) and skill differentiation (b) moderate the

negative relationship between hierarchy and performance, such that the relationship

becomes stronger as the levels of these structural characteristics increase.

Team Hierarchy Contingencies

An additional important set of contingency factors that may determine whether teams are

helped or harmed by hierarchical differentiation are aspects of the hierarchy itself. We define

hierarchy in this paper broadly as vertical differences in socially valued resources. However,

hierarchies may vary widely in the basis and structure of these vertical differences, and some

forms and bases of hierarchy are more likely to be contested than others. When hierarchies

motivate and/or allow for more individual mobility up or down the team hierarchy, such settings

increase individuals’ concerns with their self-interest (Maner & Mead, 2010; Wright, Taylor, &

Moghaddam, 1990) and the likelihood of hierarchy impairing team performance via hierarchical

struggles and conflicts (e.g., Greer & Van Kleef, 2010).


Therefore, we focus here on two critical dimensions of hierarchy which measure the

degree to which hierarchies are more likely to be contested or not – hierarchy mutability and

hierarchy form. Hierarchy mutability is defined as the degree to which the basis of a hierarchy

(e.g., status or power) is open to be contested and has been shown to increase a team’s conflict

susceptibilities (Hays & Bendersky, 2015). When hierarchies are themselves easily changed,

such as informal or status-based hierarchies, members may see more possibility for upward

movement, and lower ranked members may seek to climb up, while higher ranked members seek

to maintain rank, leading to rank-based conflicts between members (cf. Greer et al., 2017).

Additionally, in such mutable hierarchies, more disagreements about hierarchical rank may exist,

which can contribute to deteriorated individual contributions and performance-detracting

intragroup conflicts (e.g., Greer, Caruso, & Jehn, 2011; Kilduff, Willer, & Anderson, 2012).

Hierarchy mutability has indeed been able to explain when hierarchies are more likely to lead to

performance-detracting conflicts (e.g., Hays & Bendersky, 2015).

Hierarchy form is the degree to which members within the team’s hierarchy are arrayed

in an inequitable, centralized, or acyclical manner (Bunderson et al., 2016). Hierarchies whose

forms evoke more perceived inequities and politics – i.e., steep and/or centralized hierarchies,

have been suggested to evoke more conflict in hierarchical teams (e.g., Eisenhardt & Bourgeois,

1988; Harrison & Klein, 2007) than hierarchies whose form is more legitimized through

accepted downwards flowing dyadic influence patterns (what Bunderson et al. [2016] term

‘acyclical’ hierarchies). When hierarchies are experienced as more illegitimate and political,

such as steep or highly centralized hierarchies, they are more likely to promote negative team

conflicts and hurt team performance. Hierarchical form has indeed recently been found to have

implications for the conflict-susceptibilities of hierarchies (e.g., Bunderson, et al., 2016). Based


on this work, we propose:

Hypothesis 5: Hierarchy mutability (a) and hierarchy form (b) moderate the negative

relationship between hierarchy and performance, such that the relationship becomes

stronger as mutability increases and hierarchy is based on inequality (as opposed to

acyclicity).

Method

Literature Search

In order to identify all relevant studies to include in our meta-analytic test of the

hypotheses presented here, we performed a combination of search strategies. First, we performed

searches in several online databases (e.g., ABI/Inform, Google Scholar, PsycINFO, Web of

Science) to find any articles that included team or group in conjunction with the keywords

hierarchy, hierarchical differentiation, authority differentiation, power, status, influence,

leadership, expertise, inequality, and dispersion. We also performed backward and forward

searches on key theoretical articles on the topic (e.g., Halevy et al., 2011; Magee & Galinsky,

2008) to ensure we did not overlook any relevant studies. Complementing these electronic search

strategies, we manually examined the table of contents from top-tier journals in psychology,

management, and economics (e.g., Academy of Management Journal, Administrative Science

Quarterly, Applied Economics, Journal of Applied Psychology, Journal of Management, Journal

of Organizational Behavior, Journal of Personality and Social Psychology, Organization

Science, and Strategic Management Journal) published in the past five years. Lastly, to

minimize publication bias, we scanned the programs of relevant conferences and sent emails to

the members of the related associations via their newsletters (e.g., Academy of Management, the

Society for Industrial and Organizational Psychology, Interdisciplinary Network for Groups


Research, and the Society for Personality and Social Psychology), searched for unpublished

dissertations through ProQuest, and sent requests for relevant working papers to key authors in

the field. Our literature search includes articles available up until July 2015.

Inclusion Criteria and Initial Sample

Our literature search returned a list of over 1,500 potentially includable articles. In order

to be included, studies needed to contain enough information to allow for the computation of

effect sizes and report empirical measures of hierarchy, and team mediators (conflict-enabling

states, coordination-enabling processes) or outcomes (performance, viability), or both, at the

team-level of analysis. We included papers that reported effect size estimates for hierarchy and

any one potential dependent variable (mediators or outcomes) – papers did not need to include

all possible dependent variables. In addition, the hierarchy and dependent variable measures

needed to be consistent with our definitions of our core constructs in order to be included in our

sample.

Given that we define hierarchy here as vertical differences between individuals within a

team with respect to the possession of socially valued resources or assets (cf. Harrison & Klein,

2007; Zitek & Tiedens, 2012), we included hierarchy measures that were based on different

resources which are socially valued in work teams. We draw on work on expectation states

theory (e.g., Berger, Cohen, & Zelditch, 1972; Bunderson, 2003) to identify resources most

proximally related to task performance to be the most socially valued resources in teams, and we

therefore include hierarchy measures which reflect such task-related resources, including job

titles, leadership authority, and expertise levels. While other resources that are not directly

related to the task may also give rise to intragroup hierarchy, such as gender and race

(e.g., Bunderson, 2003), we excluded these from our sample because they have not been as


commonly recognized in the literature as proxies of intragroup hierarchy. For example, gender

and race are generally operationalized in terms of variety-based diversity rather than as

separation- or disparity-based diversity, precluding them from capturing team hierarchy

(Harrison & Klein, 2007). Furthermore, some attributes, especially pay levels, are often not

publicly known within the team, and do not have the same salience or relevance that internal

team bases of hierarchy, such as status or job rank would. Given these conceptual boundaries,

valid indicators for hierarchy included (but were not limited to), negotiation BATNAs (the value

of the best alternative members had to reaching a negotiated agreement in that situation),

influence networks, manipulations of the presence of leaders or layered hierarchies or not, and

status round-robins for hierarchy.

For conflict-enabling states and coordination-enabling processes, valid indicators

included self-report, video-coded, network-measures, and archival measures of the levels of

conflict-enabling states and coordination-enabling processes occurring within the team. For

example, several studies used self-report scales of conflict-enabling states (e.g., Martins,

Schilpzand, Kirkman, Ivanaj, & Ivanaj, 2013), and for coordination-enabling processes, several

studies used self-report measures, including the ability of teams to align actions, to piece together

information and provide each other regular feedback (e.g., Greer, Dannals, & Rao, 2016;

Woolley et al., 2008). Additionally, consistent with our broad definition of conflict-enabling

states and coordination-enabling processes, we not only considered direct measures of conflict

and coordination as valid indicators, but also included measures of power struggles and

psychological safety (for conflict-enabling states) and measures of knowledge integration and

information exchange (for coordination-enabling processes). For team outcomes, valid indicators

for performance included leader ratings of team performance quality, financial earnings, and


matches won, and for team viability valid indicators included self-report measures of team

satisfaction and commitment (cf., Balkundi & Harrison, 2006). Applying these inclusion criteria

yielded a usable sample of 54 relevant studies, involving 86 effect sizes, and 13,914 teams.

Thirty-nine of these studies were published, seven were unpublished dissertations, and eight

studies involved new, unpublished data. All included studies are listed in the references section,

marked by an asterisk, and an overview of the data obtained from each study can be found in

Appendix A. For an overview of excluded studies, and the reasons for their exclusion, please see

our online supplements.

Coding Procedures

Since the majority of the studies were correlational, we used Pearson’s correlation

coefficient (r) as our effect size metric. When other metrics – such as t-tests from ANOVAs –

were reported in studies, they were converted into correlations following formulas proposed by

Lipsey and Wilson (2001). When correlations were based on negative indicators of one of the

two correlated variables (i.e., such that higher scores indicated lower values of that variable), we

reverse-coded those correlations to ensure coding consistency across correlations. When studies

reported both lagged and cross-sectional correlations with the same mediator or outcome, we

only included the lagged correlations in order to be consistent with the causal directionality

implied in our framework (cf., De Jong et al., 2016). When multiple correlations of hierarchy

with the same dependent variable were available from the same study, they were combined using

linear composites to ensure independence of observations (Schmidt & Hunter, 2015), or

averaged when the information reported in primary studies was insufficient for creating linear

composites.


When correlations were based on subjectively measured variables, we corrected the

correlations for measurement error using different reliability statistics, depending on the nature

of the measure (cf., Courtright et al., 2015; Hulsheger, Anderson, & Salgado, 2009; Wang,

Waldman, & Zhang, 2014). Specifically, for single-rater measures (e.g., team leader’s subjective

rating of the team’s performance), we used the Cronbach’s Alpha to correct study correlations.

For multi-rater measures (e.g., team members’ subjective ratings of team conflict), we used

ICC(2) to correct study correlations for rater-specific measurement error whenever they were

reported, or used ICC(1)s and F-statistics to calculate ICC(2)s ourselves (cf., De Jong et al.,

2016). If none of these aggregation statistics were reported, however, we used Cronbach’s Alpha

to at least correct for item-specific measurement error in team-level mean scores. When

reliabilities were missing for subjective measures all together, we imputed the sample-size

weighted mean reliability of all available studies (cf., Steel, 2007). When hierarchy was

experimentally manipulated or when hierarchy or team processes or outcomes were measured

objectively, their reliability was set to 1. Whenever we used linear composites to combine

multiple correlations from the same study, we used Mosier reliability composites to combine

internal consistency coefficients (Schmidt & Hunter, 2015).

While extracting ‘low inference’ data from the studies (i.e., statistics) was relatively

straightforward, more detailed coding schemes needed to be developed to code ‘high inference’

data on our moderators. Almost all of our moderating variables were coded on a continuous scale

ranging from 1 to 5, with higher scores reflecting higher levels of the variable. The only

exception was hierarchy form, which was categorically coded as (1) centralization, (2) steepness,

or (3) acyclicity. Based on the coding procedures, all moderators (across all 52 studies that

reported a hierarchy-performance correlation) were independently coded by both the first and the


third author. The analyses showed a mean Krippendorf’s Alpha of .94 across moderators (SD =

.03; Hayes & Krippendorff, 2007), suggesting high levels of inter-coder agreement. Any

discrepancies between the authors were resolved via discussion until consensus was reached.

Team Task Contingencies. We operationalize task interdependence as the extent to

which team members were able to divide the team’s task or the extent to which the team’s task

required interaction to complete it (Courtright, Thurgood, Stewart, & Pierotti, 2015; Horwitz &

Horwitz, 2007). We coded task interdependence as ‘low’ (1) when team members could divide

their work into sub-tasks and work on their own and their efforts would be pooled together for a

team outcome (e.g., Mannix, 1993), as ‘moderate’ (3) when team members worked together

closely for some but not all tasks – members at times would come together to collaborate and at

other times would go off to accomplish individual sub-tasks (e.g., He & Huang, 2011), and as

‘high’ (5) when team members had to continuously work closely together, as each member’s

contribution was reliant on other team members’ contributions, and team outcomes were the

result of the collective effort of all team members (e.g., Bunderson, 2003). This is consistent with

other meta-analyses that have examined task interdependence, such as Horwitz and Horwitz

(2007), and Joshi and Roh (2009). By capturing task interdependence in a more continuous

fashion, we were able to account for both established models of interdependence, such as pooled,

sequential, reciprocal, and comprehensive, and the complex realities of tasks that influence the

nature of interactions among team members in ongoing teams (LePine et al., 2008). To infer task

interdependence, we drew on information about the collaborative nature of work as described in

the description of the sample and team task. For example, baseball teams, where members could

function loosely from one another, were coded as having low interdependence, while

management teams, where members at times function independently in their own units and at


times interdependently on management decisions, were coded as having moderate

interdependence, and student lab teams, where members are highly dependent on one another to

accomplish the team task, were coded as having high interdependence. The studies included in

this meta-analysis reflected all levels of task interdependence.

Kahn et al. (1964) argued that task ambiguity arises when an individual lacks the

necessary information about the task to carry out his or her role. Specifically, according to Kahn

et al. (1964), three subtypes of task ambiguity can arise related to a) the scope of responsibilities

(i.e. what is expected of me?), b) the behavioural responsibilities (i.e. what activities will lead me

to accomplish what is expected of me?), and c) the hierarchy of responsibilities (i.e. what tasks

are prioritized over others?). To our knowledge, we are the first to examine task ambiguity in a

meta-analytic fashion. We therefore leveraged this definition by Kahn et al. (1964) to develop a

coding scheme for task ambiguity. We coded task ambiguity as ‘low’ (1) when members had

full, complete information about their task responsibilities within the team, including the scope,

nature, and prioritization of their task responsibilities and behaviors within the team, as

‘moderate’ (3) when members had some clarification and boundaries around what behaviors and

activities were expected of them individually on the team task, and task ambiguity was coded as

high (5) when members lacked clarity about task responsibilities and expectations, including the

scope, nature, and prioritization of member responsibilities within the team. To infer task

ambiguity, we drew on information about the goal, task, and performance expectations of the

teams studied as described in the methods sections of the papers we coded. Settings, such as

teams of healthcare specialists, where the task and associated responsibilities were clearly known

to members, were coded as having low task ambiguity (Mitchell et al., 2015). In contrast,

settings in which teams did not have much knowledge about how to accomplish the task and had


to figure it out on their own, such as startup teams, where teams had to figure out who would do

what and what needed to be done to reach a good decision, were coded as having high task

ambiguity (Foo, Sin, & Yiong, 2006). When samples were composed of a mixture of teams, for

instance due to snowball sampling, samples were coded as having moderate task ambiguity (Van

der Vegt et al., 2010). Included studies reflected all levels of task ambiguity.

In contrast to task ambiguity, in which team members lack information about their role

and how to prioritize their responsibilities, task complexity implies that members do have

information about the task, but that the information load is high. Namely, task complexity arises

when there is the presence of multiple pathways to meet a goal, the presence of multiple desired

outcomes, the presence of conflicting interdependence among paths to multiple outcomes, and

the presence of probabilistic links among paths or outcomes (Campbel, 1988). In short, task

accomplishment is quite detailed and intricate, and in need of considerable thought and

exploration. In coding task complexity, we leveraged this definition to develop a coding scheme.

We coded task complexity as ‘low’ (1) when the team’s task was simple and a limited number of

task activities were clearly linked to outcomes, as ‘moderate’ (3) when there were a few different

goals or paths to the goals, but the task was relatively straight-forward, and task complexity was

coded as ‘high’ (5) when numerous goals, and/or paths to the goals existed, and considerable

thought was required to achieve desired outcomes. To infer task complexity, we relied on the

descriptions of the team’s task. Tasks, such as regularly scheduled team meetings, were coded as

low in complexity due to their regular occurrence (Sauer & Kauffeld, 2013). Tasks, such as those

of a top management team, were coded as high in complexity due to the presence of numerous

subgoals and strategies to task accomplishment (Boone & Hendriks, 2009). All levels of task

complexity were present in the dataset, with a larger number of moderately low complex tasks.


Team Structure Contingencies. Membership instability is one of the two factors that

make up a team’s temporal stability, in addition to team longevity (Hollenbeck et al., 2012). Our

coding scheme for membership instability was based on Lee, Koopman, Hollenbeck, Wang, and

Lanaj’s (2015) team descriptive index short form. Lee and her colleagues developed a 9-point

Thurstone response scale with 5 descriptive anchors. We reversed and compressed the 9-point

scale to a 5-point scale by retaining the anchors for each of the levels and narrowing the focus of

the descriptors to only reflect membership instability. Membership instability was coded as ‘low’

(1) when teams experienced very little membership change or turnover resulting in team

members that were very familiar with each other and a team that rarely, if ever, recruited new

members, as ‘moderate’ (3) when teams had some level of membership change, but whereby

there was a main, core set of team members, and as ‘high’ (5) when teams had high levels of

membership change and members were almost total strangers to each other or did they spend

enough time together to really get to know one another. To infer membership instability, we

drew on information on the brevity of the team’s time together and the permeability of team

boundaries. Teams at all levels of membership instability were included in this meta-analysis.

Management teams (Haleblian & Finkel, 1993), manufacturing teams (Sauer & Kauffeld, 2013),

healthcare teams (Thylefors, 2012), semester-long student project teams (Martins et al., 2013),

and one-shot lab teams (Tost et al., 2013) are examples of teams coded from low to high at each

level of membership instability, respectively.

In the same paper, Lee et al. (2015) developed a 9-point Thurstone measure for skill

differentiation with five descriptive anchors. Skill differentiation describes the extent to which

members have specialized knowledge or functional capacities (Hollenbeck et al., 2012). Similar

to our procedure for membership instability, we compressed the 9-point format to a 5-point scale


by retaining the anchors for each of the levels. Skill differentiation was coded as low (1) when

team members’ roles, responsibilities, and skillsets were overlapping and/or highly substitutable

with one another or the tasks were so simple that even untrained people could perform almost all

of the tasks, as ‘moderate’ (3) when members had a core set of skills, roles, and responsibilities

that they all shared but each member still had some degree of unique skills and contribution to

the task, and as ‘high’ (5) when the team had a clear lateral division of labor, based on

differences in members’ unique roles, responsibilities, and areas of expertise. The extent of skill

differentiation was relatively evenly distributed across studies in our sample. The following are

examples of teams ranging from low to high, respectively: Student lab teams (e.g., Lemoine,

Koseoglu, & Blum, 2015), negotiation teams with different role instructions (Pinkley, Neale, &

Bennett, 1994), R&D teams (Ren, Gray, & Harrison, 2015), and top management teams

(Hambrick, Humphrey, & Gupta, 2015). To infer skill differentiation, we drew on information

about the degree to which different specializations were present in the team.

Team Hierarchy Contingencies. Hays and Bendersky (2015) recently found that the

extent to which team members perceive that they can change their position in the hierarchy

drives more competitive and conflict prone behaviors in teams. They coined the term hierarchy

mutability to describe the extent to which the basis for the team’s hierarchy allowed for the

opportunity for upward mobility. We coded a team’s hierarchy mutability as ‘low’ (1) when

team members were unlikely to be able to change their position in the team hierarchy, such as in

the teams of medical clinic nurses examined by Perry (2014), in which job positions are

generally stable. Teams were coded as ‘moderate’ (3) when team members had some potential to

change their position in the hierarchy, such as the production teams examined by Bunderson

(2003) which allowed for some potential in the change of the hierarchical ranking of members.


Finally, teams were coded as ‘high’ (5) in hierarchy mutability when team members could easily

change their position in the team hierarchy, such as the start-up teams examined by Foo et al.

(2006) which operated in a dynamic and complex environment in which team members were

trying to find their roles. To infer hierarchy mutability, we drew on information on the

entrenchment and type of hierarchy described in the study. While teams were identified at all

levels of mutability, the distribution in our sample was clearly bimodal with the majority of

teams scoring either toward the lower (1) and (2) or the higher (5) hierarchy mutability values.

Bunderson et al. (2016) argue that the way the hierarchy is structured matters for its

outcomes. Following their reasoning, we coded hierarchy form in four different categories.

Following Bunderson and colleagues, we distinguished between (1) centralization, (2) steepness,

(3) acyclicity, and (4) mixed. The centralization approach conceptualizes hierarchy as the

concentration of hierarchy characteristics (e.g., power, status) in one team member or a small

group of team members. Centralization is captured through measures of concentration, such as

the coefficient of variance or the Gini coefficient. Examples of scholars studying this type of

hierarchy form are He and Huang’s (2011) measure of board membership inequality and Rulke

and Galaskiewicz’ (2000) centralization of the teams’ network structures. The steepness

approach conceptualizes hierarchy as the aggregate difference across members on hierarchy

characteristics. Steepness is captured through measures of standard deviation and Euclidian

distance. Examples of this type of hierarchy form are the standard deviation of power levels that

Greer and Van Kleef (2010) used and the manipulation of the leadership role that Becker and

Baloff (1969) used to study the effectiveness of different organizational structures. The most

recently established form of hierarchy, acyclicity, is conceptualized as “cascading relations of

dyadic influence (p. 1270)”. Acyclicity is operationalized as a network of directed influence


relations, for instance, by calculating Krackhardt’s (1994) hierarchy index. Given the recent

developments with regard to this form of hierarchy, we were able to uncover only two studies

that operationalized hierarchy as acyclicity in terms of task-relevant resources, as we focused on

here, and looked at group-level processes and outcomes (Bunderson et al., 2016; Cantimur, Rink,

& Van der Vegt, 2015). Given the categorical nature of our coding scheme, we coded hierarchy

form as ‘mixed’ when studies reported multiple forms of hierarchy.

Meta-Analytic Approach

We performed our meta-analysis using Version 3 of the Comprehensive Meta-Analysis

software (Borenstein, Hedges, Higgins, & Rothstein, 2014), which relies on the Hedges and

Olkin (1985) approach. Consistent with best practice recommendations (Geyskens, Kirshnan,

Steenkamp, & Cunha, 2009), we used a random effects model, which assumes that variability in

correlations across studies is caused by both sampling error and true effect size variability in the

population (Schmidt & Hunter, 2015).

There is growing scholarly recognition that, rather than interpreting effect size

magnitudes in absolute terms using universal standards (e.g., Cohen, 1988), effect size

magnitudes should be interpreted in reference those found in prior research in related areas

(Bosco, Aguinis, Singh, Field, & Pierce, 2015; Paterson, Harms, Steel, & Crede, 2016).

Consistent with this recommendation, we relied on the average effect size estimate across team-

level meta-analyses (r = |.26|) reported by Paterson et al. (2016) as our reference for interpreting

our main effects, and the average effect size estimate across team-level meta-analyses (semi-

partial correlations: sr = |.20|) reported by O'Boyle, Banks, Walter, Carter, and Weisenberger

(2015) for interpreting our moderator effects. In absence of pre-existing benchmarks for

mediation effect sizes, we constructed a benchmark ourselves by identifying seven prior team-


level meta-analyses that tested mediation effects (e.g., Knight & Eisenkraft, 2015), extracting the

reported indirect effect size estimates for each meta-analysis (see online supplements), and then

using the reported harmonic means to calculate the sample-size weighted average across those

meta-analyses (marked by double asterisks in the reference section). This resulted in a mediated

effect size benchmark of ab = |.09|. Accordingly, we interpret our effect sizes in reference to the

above effect size estimates, and qualify them as ‘below-average’, ‘average’ and ‘above-average’

(cf., De Jong et al., 2016).

Besides interpreting the magnitude of our effect size estimates, we also consider the

precision with which these estimates capture the true effect in the population by constructing

95% confidence intervals (CIs) around these estimates. A 95% CI represents an estimation of the

range of plausible values of the true (unknown) population parameter, and indicates that we can

be 95% confident that the interval’s lower and upper bounds contain that population parameter

(Cumming, 2012; Ellis, 2010). The narrower the CI, the more precise the point estimate is in

capturing the true population parameter. CIs can also be interpreted as indicators of statistical

significance (Cummings, 2012). Specifically, a 95% CI that does not include zero may be

interpreted as a statistically significant result at the level of p < .05.

As meta-analytic findings can be severely compromised by publication bias, we first

assessed the presence of such bias in our data using multiple techniques, namely the Begg and

Mazumdar’s rank correlation test, the Egger’s intercept test, and Duval and Tweedie’s (2000)

trim-and-fill method (for more details of each test, see Kepes, Banks, McDaniel, & Whetzel,

2012). The Begg and Mazumdar’s rank correlation test was non-significant (t = -.16, p = .09), as

was Egger's regression intercept test (b0 = -.19, p = .59). Furthermore, both the Duval and

Tweedie’s Trim and Fill method and associated funnel plot did not indicate skew in the


distribution of effect sizes, and/or the need to add more effect sizes on any side of the

distribution to make it more symmetric. Together, these tests consistently suggest that

publication bias was not a major concern in our sample.

Results

Mediation Analysis

Although a main effect between hierarchy and effectiveness is not a requirement for

mediation to exist (Aguinis, Edwards, & Bradley, 2017), we nonetheless assessed this overall

main effect in order to gain insight into whether the positive or the negative effect of hierarchy is

more likely to win out at this point in the literature. Our results show a net negative point

estimate in relation to team performance (r = -.08). The CI indicates that the true main effect

size is unlikely to be zero (CI95% = -.13, -.04), indicating the presence of a negative effect of

hierarchy on performance. Likewise, the findings also indicate a negative effect size estimate in

relation to team viability (r = -.11), with the CI suggesting a true effect size is highly likely to be

below zero despite the non-trivial uncertainty surrounding the point estimate (CI95% = -.19, -.02).

Thus, we find that, at this point in the literature, hierarchy, on net, negatively impacts both team

performance and viability.

We subsequently proceeded to analyze the mediated effects of hierarchy using meta-

analytic structural equation modeling (MASEM) procedures outlined by Viswesvaran and Ones

(1995). The first step in this analysis involves estimating true mean correlations for all pairs of

variables that make up the conceptual model, and pooling these correlations into a matrix. To

populate this matrix, we used the afore-mentioned Hedges and Olkin approach with random

effects to calculate mean corrected correlations of hierarchy with each of the mediators and team

effectiveness outcomes. Correlation estimates among the mediators and team effectiveness


outcomes were obtained from prior meta-analyses conducted by LePine and colleagues (2008)

and Maynard, Mathieu, Gilson, O'Boyle, and Cigularov (2013). Pooling these meta-analytically

derived correlations yielded the matrix displayed in Table 1. In the second step, we specified our

mediation model – with hierarchy as the independent variable, coordination-enabling processes

and conflict-enabling states as mediators, and team performance and viability as dependent

variables – and used the correlation matrix as input to test this model in LISREL 9.1 (Jöreskog &

Sörbom, 1996). Given that sample sizes differed across meta-analytical correlations, we used the

harmonic mean as our sample size. Subjecting the correlation matrix to LISREL and testing our

multiple mediation model yielded beta-coefficient estimates, standard errors, and CIs for each of

the individual paths. We subsequently used the RMediation package (Tofighi & MacKinnon,

2011), which relies on the distribution-of-the-product method, to build 95% CIs around the

mediated effect (cf. Courtright et al., 2015). Hypothesis 1 predicted that coordination-enabling

processes would mediate the positive relationship between hierarchy and our two team

effectiveness outcomes. Contrary to Hypothesis 1a, the results shown in Table 2 indicate that

while coordination-enabling processes do mediate the hierarchy-performance relationship, and

the mediated effect estimate is below-average in magnitude, the point estimate is negative rather

than positive (ab = -.04), with a negative path running from hierarchy to coordination-enabling

processes (β = -.17) and a positive path running from coordination-enabling processes to

performance (β = .21). The CI indicates that while the there is a non-trivial degree of imprecision

around the mediated effect estimate, the true mediated effect size parameter is likely to be

negative (CI95% = -.05, -.02), and not positive as we hypothesized.

The same pattern of results was found for team viability. The findings indicate a below-

average mediated effect of hierarchy on viability via coordination-enabling processes, but the


effect is negative rather than positive in nature (ab = -.04), with a negative path running from

hierarchy to coordination-enabling processes (as already indicated above) and a positive path

running from coordination-enabling processes to viability (β = .23). The CI indicates that the true

mediated effect size is likely to be negative (CI95% = -.06, -.02), instead of the positive indirect

effect we predicted. Thus, our results also fail to support Hypothesis 1b.

Hypothesis 2 proposed conflict-enabling states as a mediator of the negative relationship

between hierarchy and team effectiveness. In support of Hypothesis 2a, we find a negative

mediated effect estimate of hierarchy on performance via conflict-enabling states (ab = -.02),

based on a positive path running from hierarchy to conflict-enabling states (β = .17) and a

negative path running from conflict-enabling states to performance (β = -.13). The results also

indicate that the true mediated effect parameter is below-average, but despite some imprecision

in this estimate, it is likely to be negative (CI95% = -.04, -.01), indicating the presence of a

negative indirect effect and support for Hypothesis 2a. Similarly, our results also show a negative

indirect effect estimate of hierarchy on viability (ab = -.03), with a positive path from hierarchy

to conflict-enabling states (as already indicated above) and a negative path from conflict-

enabling states to viability (b = -.18). The effect is below-average, but the CI indicates that we

can be confident that the true mediated effect size is likely to be negative (CI95% = -.05, -.02).

These results provide support for Hypothesis 2b.

Moderation Analysis

To assess whether moderator analysis was warranted, we first assessed effect size

heterogeneity across primary studies using two complementary indices: the Q-statistic and I2

(Higgins & Thompson, 2002; Huedo-Medina, Sánchez-Meca, & Botella, 2006). Our analysis

across the 52 effect size estimates for the hierarchy-performance relationship in our sample


showed the presence of considerable variation in effect sizes (Q = 176.97, p = .00) and suggests

that most of this variation may be explained by between-study differences (I2 = 71.18). We

therefore proceeded to test our moderator hypotheses (H3-H5) by applying random effects,

weighted least squares (WLS) meta-regression, which allows for examining the influence of

continuous – as opposed to discrete – moderator variables, and examining the influence of

multiple moderator variables simultaneously – as opposed to one at a time (Steel & Kammeyer-

Mueller, 2002). While WLS meta-regression yields unstandardized beta coefficient estimates for

each of the moderators (B), the effect size magnitude benchmark we rely on is expressed as a

semi-partial correlation (sr). We therefore also converted our results into semi-partial

correlations, using the formula provided by Cohen and Cohen (1983), in order to interpret the

effect size magnitudes of our moderators in light of this benchmark.

Team Task Contingencies. We proposed in Hypothesis 3 that a) task interdependence;

b) task ambiguity; or c) task complexity would moderate the positive relationship between

hierarchy and team performance, such that the relationship would become stronger as the levels

of these task characteristics increase. The results of meta-regression analysis failed to support

Hypothesis 3a (see Table 3). While the coefficient estimate for the relationship of task

interdependence with the hierarchy-performance correlation is above-average in magnitude (sr =

-.38), it is negative rather than positive in nature (B = -.07). The CI around the estimate confirms

that the true parameter coefficient is likely to be negative as opposed to positive (CI95% = -.12, -

.03). These results thus contradict Hypothesis 3a. Hypothesis 3b regarding task ambiguity was

supported. The results show a positive, above-average coefficient estimate for the relationship

between task ambiguity and the correlation between hierarchy and team performance (B = .14; sr

= .45), and suggest that we can be 95% confident that there is a positive, non-zero true


moderation effect in the population (CI95% = .07, .21). Hypothesis 3c regarding task complexity

was not supported. The results yielded a positive effect (B = .03), but the estimate was below-

average (sr = .11) and surrounded by relatively high levels of imprecision that include zero

(CI95% = -.03, .08).

Team Structure Contingencies. Hypothesis 4 predicted that a) membership instability

or b) skill differentiation would moderate the negative relationship between hierarchy and team

performance, such that the relationship becomes stronger as the levels of these structural

variables increase. As shown in Table 3, our results show support for Hypothesis 4a:

membership instability was negatively related to the correlation between hierarchy and team

performance (B = -.05), and the point estimate is above-average in magnitude (sr = -.25). While

the upper limit of the CI approximates zero (CI95% = -.10, -.004), it does not include zero,

indicating that while the estimate is surrounded by relatively high levels of imprecision, the true

moderation effect is likely to be negative. The results also revealed support for Hypothesis 4b,

showing a negative coefficient estimate for the relationship between skill differentiation and the

correlation between hierarchy and team performance (B = -.09). The estimate is above-average in

magnitude (sr = -.36) and the CI indicates that the true moderation effect parameter is likely to

be negative (CI95% = -.15, -.03). These findings offer support for Hypothesis 4.

Team Hierarchy Contingencies. We proposed in Hypothesis 5 that a) hierarchy

mutability or b) hierarchical form would moderate the negative relationship between hierarchy

and performance, such that the relationship becomes stronger as mutability increases and

hierarchy is based on inequality (as opposed to acyclicity). In support of Hypothesis 5a, we find

a negative relationship between hierarchy mutability and the hierarchy-team performance

correlation (B = -.11), with an above-average effect size magnitude (sr = -.35), and CIs that


suggest we can be confident that the true moderation effect in the population is negative (CI95% =

-.19, -.04). Thus, Hypothesis 5a is supported. Contrary to Hypothesis 5b, the results indicate a

negative coefficient estimate for acylical hierarchies (B = -.48) that was above-average in

magnitude (sr = -.30), suggesting that for such hierarchy forms the relationship between

hierarchy and performance is considerably more negative than for inequitable hierarchies

(whether steep or centralized). To gain further insight, we disaggregated inequality into two

separate hierarchy forms (steepness and centralization) and employed a ‘shifting unit of analysis’

approach (Cooper, 2010), which involves computing separate correlations for each hierarchy

form when hierarchy forms were reported within the same study (thereby increasing the number

of acyclicity studies to two). We then re-analyzed our regression model twice, once using

steepness and once using centralization as the reference category. These analyses revealed that

the population coefficient for acylical hierarchies was likely to be considerably more negative

than that for steep hierarchies (B = -.28; CI95% = -.52, -.04), but unlikely to be different in

magnitude from that of centralized hierarchies (B = -.18; CI95% = -.42, .06). While the CIs

suggest a true difference in effect size between these hierarchy forms is likely to be present in the

population (CI95% = -.83, -.12), the result was based on a restricted sample of only two studies

containing an acyclicity measure, making it hard to confidently interpret these results.

Irrespectively, Hypothesis 5b is not supported.

Together our seven moderators explain 43% of the variance in hierarchy-performance

correlations across studies.

Supplementary Analyses

Hierarchy Measurement. Because the ways in which the authors in the hierarchy

literature measure hierarchy vary quite dramatically, we examined whether hierarchy’s effects on


performance varied based on how hierarchy was measured. As seen in Table 4, the effect sizes

do not markedly differ across the different measurements of intragroup hierarchy. Hierarchies

measured on differences in expertise levels appear to be the most positive for team performance,

while hierarchies based on formal positions in a team appear to be the most negative. However,

given that these estimates are based on relatively small sample sizes, they may be susceptible to

second-order sampling error (Schmidt & Hunter, 2015). We therefore caution readers to avoid

over-interpreting potential differences across hierarchy measures reported in Table 4.

Causality. In our sample, some studies were causal (experimental) while others were

correlational (field) in nature. To explore whether our findings across this broad sample are on

par with only looking at the effects from purely causal studies, we checked whether the findings

differed between the correlational and experimental studies using WLS meta-regression. We did

not find meaningful differences between experimental studies (the reference category) and

correlational studies in the effect of hierarchy on team effectiveness (B = -.02, CI95% = -.07, .11).

Mediation Analyses. To gain further insight into the generalizability of our mediation

results, we complemented Viswesvaran and Ones’s (1995) fixed effects MASEM approach with

a random effects (FIMASEM) approach recently developed by Yu, Downes, Carter, and O'Boyle

(2016). This approach uses both meta-analytical correlations and standard deviations (reported in

Table 1) to generate a large number of bootstrap samples on which the proposed mediation

model is tested, thereby providing insight into the effect size heterogeneity associated with

individual path coefficients and into the percentage of samples that yield an acceptable fit of the

model as a whole. In short, the results of the FIMASEM analysis show path coefficients for

individual and mediated paths that are directionally consistent with those from the fixed effects

approach, as would be expected. More importantly, though, they also indicate that both the


individual path estimates and the negative dual-pathway model as a whole generalize to a

majority of the bootstrap samples. These results thus provide further support for Hypotheses 2a-b

and further evidence against Hypotheses 1a-b. More details can be found in the online

supplements.

Moderation Analyses. While the advantage of WLS regression is that it accounts for

shared variance among the moderators, estimating seven moderators simultaneously with only 52

studies could potentially produce unstable estimates due to statistical power issues. Indeed, this

might explain why some of the findings did not support our hypotheses. To examine this

possibility, we re-ran our moderator analysis twice, once for the subset of moderators that did

support our hypotheses (i.e., task ambiguity, membership instability, skill differentiation,

hierarchy mutability) and once for the subset of moderators that did not (i.e., task

interdependence, task complexity, hierarchy form). This approach helps overcome issues of

statistical power by increasing the moderator-to-study ratio, while still accounting for shared

variance by including multiple moderators within each test. These analyses produced results very

similar to our original analyses, in that they also supported the moderating roles of task

ambiguity, skill differentiation, and hierarchy mutability, and also failed to support the

moderating roles of task interdependence, task complexity, and hierarchy form. The only

exception was the result for membership instability: the moderating effect supported in our

original analysis was not supported when re-running the analysis on the moderator subset. Taken

together, the consistent results strengthen confidence in the results from our original analysis and

the stability of the estimates for nearly all of our moderators, including those that did not support

our hypotheses. These analyses can be obtained from the second author.

Discussion


Our investigation of why and when hierarchy impacts team effectiveness represents a first

conceptual and empirical integration of the different perspectives on hierarchy in teams,

including the functionalist, conflict, and contingency perspectives. Together, the key variables

and predictions associated with these perspectives and our empirical findings in this meta-

analysis are indicative of an emergent, integrative conceptual framework of the way hierarchy

operates in team settings (see Figure 1). We recognize that our meta-analytic approach is limited

in the sense that our model is derived from an aggregation of ideas and results presented in prior

research. We therefore highlight here how our review and findings are indicative of a theory of

hierarchy in teams which can be more fully and robustly tested through future primary studies

that examine the set of relationships we suggest may categorize the broader literature here.

Specifically, as the figure illustrates, in the field of research on hierarchy in teams,

functionalist and conflict perspectives have painted very different pictures of the effects of

hierarchy on team outcomes (positive versus negative). Additionally, different suggestions have

arisen on the upstream contingencies which can reconcile these perspectives, and determine

whether hierarchy sets into play performance-detracting conflict states or performance-

enhancing coordinated action processes. By examining the upstream contingencies which

determine the types of processes hierarchy sets into play in teams, researchers can develop the

most targeted interventions to course-correct the effects of hierarchy at their root. Our meta-

analytic examination provides a first direct comparison of the mechanisms and moderators

proposed within these different perspectives, and our findings suggest that several key aspects of

these perspectives portrayed in our figure are in need of further qualification and extension in

future primary studies. For instance, our findings challenge the strength of the functionalist

perspectives, add nuance to the conflict perspectives by identifying amplifying or ameliorating


contingencies, and parse out the most potent moderators from the less potent ones. These

qualifications and extensions have important implications for the hierarchy literature moving

forward, which we elaborate on in more detail below.

Theoretical Implications and Avenues for Future Research

The co-existence of the functionalist and conflict perspectives on hierarchy has given rise

to the theoretical notion that hierarchy might act as a double-edged sword, capable of both

helping and harming teams (Anderson & Willer, 2014). Our meta-analysis provide a first

empirical test of this notion, and our results offer very little support for the functional

perspective, but overwhelming support for the conflict perspective, showing an overall negative

main effect of hierarchy across team effectiveness outcomes, a negative mediating effect of

conflict-enabling states that generalizes across the majority of our bootstrapped samples, and

support for three of the four moderators associated with the conflict perspective with above-

average effect size magnitudes. This evidence suggests that, at this point in the literature, one

edge of the sword’s blade – i.e., the conflict perspective – is considerably sharper than the other,

and that there is a need to reconsider past assumptions on the functionality of hierarchy. Indeed,

while the functionalist perspective has received much more theoretical attention, and is held as a

common popular belief, our findings directly challenge this belief, suggesting that this belief

may have been over-extended from just a few studies that found support for the positive

mediating role of coordination (i.e., only three of the 54 studies in our meta-analysis found

support for this: Boone & Hendricks, 2009; Halevy et al., 2012; Woolley et al., 2008), or from

studies that did not directly test the impact of hierarchy on team task performance (i.e., Halevy et

al., 2011; Tiedens & Fragale, 2003; Zitek & Tiedens, 2012). Our findings do not imply that

hierarchy cannot be functional for teams, but they do suggest that the positive coordination


pathway proposed in the literature is of limited help in explaining why hierarchy enhances team

effectiveness, and that there is need for future research to critically rethink the current

functionalist perspective in the literature. For example, it may be that hierarchy generally is

functional for team effectiveness, but may only spur the performance of certain individuals

within the hierarchy (i.e., those on top).

By contrast, while the conflict perspective has, up until recently, received far less

attention in the literature (for a recent review of this perspective, see Greer et al., 2017), our

meta-analytical evidence suggests that this perspective offers a particularly potent explanation

for how hierarchy operates in teams, and we extend thinking in this literature by incorporating

the different mechanisms under the grouping of conflict-enabling states and by linking the

theoretically connected moderators under the heading of conflict-susceptibilities. Our findings

suggest that this perspective, and the extensions on it we make here, should receive more

systematic attention in the future. In particular, we believe more work should be done in

extending our understanding of the negative and nefarious ways in which hierarchy can drive

performance-detracting conflicts in teams. For one example, research might focus on unpacking

the micro-mediating mechanisms underlying the relationship between hierarchy and conflict, and

identifying contingencies or interventions best suited to mitigate these mechanisms. To illustrate,

consider Van Bunderen et al.’s (2017) suggestion that hierarchy relates to conflict because it

leads members at different ranks to have different needs and interests. This suggests that settings

which amplify different individual interests should bring out the most negative effects of

hierarchy, and indeed we find that when skill differentiation is high (which should amplify

members having different interests), hierarchy is most negative for performance. Interventions

which can align members around collective interests, such as group-based rewards, would be


interesting to investigate in future research as one potential extension of the conflict perspective.

As another example, we adopted a broad view of conflict-enabling states and coordination-

enabling states in an effort to capture as many of the disparate mediators as we could and to

provide an integrative look at the different positive or negative explanations of why hierarchy

impacts team effectiveness. However, regarding the negative explains for instance, we do believe

that hierarchy most likely relates to conflicts over ranks (i.e., status conflicts, Hays & Bendersky,

2015; power struggles, Greer & Van Kleef, 2010) or resources (i.e., process conflicts about the

allocation of resources, De Wit, Greer, & Jehn, 2012). While data limitations did not allow us to

test this idea, future research would therefore benefit from examining these specific types of

conflict states in order to better understand whether the effects of hierarchy are bound to a certain

form of conflict-enabling state, or generalize across all conflict-enabling states.

In addition to having implications and extensions for the broader functionalist and

conflict perspectives on hierarchy, our results also extend emerging contingency perspectives

(e.g., Anderson & Willer, 2014; Bunderson et al., 2016) on when hierarchy impacts teams.

Specifically, in showing that some contingency factors can explain when hierarchy is particularly

beneficial (e.g., high task ambiguity) while others explain when it is particularly harmful (e.g.,

high temporal instability), our findings suggest that contingency factors complement each other

in explaining how hierarchy can be both helpful and harmful to teams. These findings point to

the need to examine moderators in a more integrative manner, and shift attention to considering

their joint impact, in order to more fully understand how hierarchy operates in teams.

Conversely, our findings also imply that, in singling out individual moderators, prior research

may have under- or overestimated the true impact of hierarchy within their particular research

context (e.g., Bunderson et al., 2016; Ronay et al., 2012), and that they might have arrived at


different or more nuanced conclusions if they had simultaneously accounted for multiple

contingencies that might have been at work. As such, our integrative analysis extends the

contingency approach advocated by others (e.g., Anderson & Willer, 2014; Bunderson et al.,

2016) by pointing out the need for considering integrating multiple complementary

contingencies at a time. Specifically, in light of the limited evidence for the functional

contingencies and convincing support for the conflictual contingencies, we identify

understanding when the former outweigh the latter as an important direction for future work.

Beyond the general recommendation of examining multiple contingency factors, our

findings also are suggestive of which moderators should be included in future research by

providing insight into which moderators are the most powerful (i.e., task ambiguity, skill

differentiation, membership instability, and hierarchy mutability) and which are the least (e.g.,

task complexity). The fact that the most powerful moderators offer unique predictive validity

above and beyond the other moderators in our model attests to the robustness of their impact. For

example, our results attest to the robustness of Hays and Bendersky’s (2015) recent finding on

hierarchy mutability, suggesting that when hierarchies are more mutable, they are more likely to

be contested and harm performance. Similarly, our results on team structures susceptible to

conflict (high membership instability and/or skill differentiation) support theorizing by

Gruenfeld and Tiedens (2010) that suggests that hierarchy and homogeneity are most effective

when combined, and that teams can descend into silos and conflicts when only vertically or only

horizontally differentiated. These first three moderators, and the robust support for them, can be

explained in part due to their centrality within the conflict perspective – when teams have

structures such as these, which leave teams easy pretty to conflict, hierarchies are likely to

backfire for team performance. We did find one moderator which offered a ray of hope for


functional accounts of hierarchy – task ambiguity. By showing that when tasks are highly

unclear, hierarchy does prove to be beneficial, task ambiguity may provide a useful lens to better

understand how and why hierarchy could still be useful for teams, such as by bringing order to

chaos and reducing uncertainty.

For the less supported moderators in our meta-analysis, it suggests that the effect size

magnitude and direction of some moderators examined in past research may have been

overstated or misrepresented. Notably, contrary to Bunderson and colleagues (2016), we actually

found that acyclical hierarchies were more negative for team performance than inequitable

hierarchies. We note that this finding should be interpreted with caution, as this comparison was

based on only a small number of studies using acyclical hierarchy measures. Nonetheless, our

results clearly signal the need to re-consider and explore how different forms of hierarchy, such

as the forms introduced by Bunderson et al. (2016) as well as other facets of hierarchy, may

differentially shape the hierarchy-performance relationship. Our predictions regarding task

complexity were not supported, and for task interdependence, we even found that it enhanced the

team performance detriments of hierarchy. In retrospect, the latter finding might be understood

in light of conflict theories– in loosely interdependent teams, members may not interact as much

and thereby have less chance for conflict. In such situations, hierarchies may be less contested

and more likely to help performance though as this literature progresses further work identifying

the contingencies that elicit positive effects of hierarchy will be helpful.

Together, our findings challenge past contingency perspectives which have focused on

single moderators, and extend past work by showing which combination of moderators may

work best in combination – task ambiguity, which can explain the bright side of hierarchy, and

conflict-susceptibility moderators, such as membership instability, skill differentiation, and


hierarchy mutability. These findings challenge the strength of the functionalist perspective, and

add nuance to the conflict perspective by identifying the contingencies most relevant to

amplifying or ameliorating the more negative effects of hierarchy in teams and thus help us

interpret how to move forward in theorizing about the nature and effects of hierarchical

differentiation in teams.

Methodological Implications

Our meta-analysis and emergent theory also offer considerations for the methodological

designs of future primary studies. First, as we depict in our theoretical model, past research has

proposed that hierarchy can benefit performance via coordination or hurt performance via

conflict. When we incorporate both of these pathways in one model, we find that the negative

pathway wipes out any benefits from the positive pathway. This suggests that future empirical

studies would benefit from simultaneously examining both the positive as well as negative

pathways by which hierarchy affects performance to understand the net overall effects hierarchy

will have on performance. This could include examining conflict and coordination, or conflict

and other potentially more powerful positive forces by which hierarchy could impact

performance, such as shared mental models or role clarity.

Second, and relatedly, we discuss the benefits here of examining multiple moderating

factors in conjunction, as this allows the direct comparison of different contingency factors.

Because our conclusions are limited to inference from others’ primary data here, we encourage

future research to continue to compare the predictive ability of variables from the two different

moderator categories that we induce in our model here (task coordination demands and intra-

team conflict susceptibilities). For example, research could compare whether task complexity or

intra-team diversity better determine whether hierarchy helps or hurts team performance.


Third, when designing future primary studies, we believe multi-method research is

important. We did not find large differences here based on how hierarchy was measured, and

especially not in lab or field differences, and continuing to examine hierarchy in both lab and

field settings can help rule out potential concerns about both generalizability in lab studies as

well as third variable problems in field studies.

Fourth, when designing future primary studies, we note that while there do not appear to

be large effects based on the type of hierarchy being measured, some hierarchies (i.e., expertise)

may have more positive effects than others (i.e., positional). Therefore, being intentional about

the base of hierarchy one measures and making sure the theory is explicitly tied to the base being

measured is important and may yield more nuanced and precise knowledge in this area as we go

forward. We noted that historically many papers would not align the base of hierarchy being

measured with the theory, theorizing about status hierarchies and measuring influence

hierarchies, or theorizing about power hierarchies and then measuring positional hierarchies

which may reflect both power and status differences in teams. More precision in construct clarity

thus can help the field move forward.

Limitations and Future Opportunities

In addition to setting out future areas in need of conceptual attention in future research,

our meta-analysis also highlights issues to keep in mind in future empirical hierarchy research.

First, while the effect size estimates for the supported moderators are all above-average, the

effect sizes for our main and mediated effects were below-average. There are several reasons

why this might be the case. First, compared to proximal outcomes (e.g., emergent states,

behavioral processes), team performance represents a very distal outcome to team hierarchy that

is more likely to be impacted by exogenous factors that are not under the direct control of the


team (e.g., environmental rivalry, turbulence, etc.; Beal, Cohen, Burke, & McLendon, 2003), and

therefore typically yield smaller effect size magnitudes (De Wit et al., 2012). Indeed, our

findings regarding impact of hierarchy on the two mediators versus performance are consistent

with this pattern. Second, the vast majority of the team performance measures were measured

objectively (75% of studies) rather than subjectively, which tends to result in smaller effect size

magnitudes due to a combination of reduced bias and decreased content adequacy (Mesmer-

Magnus & DeChurch, 2009). A third reason is that there was considerable effect size

heterogeneity across the primary studies in our sample, which dampened the overall mean effect

size estimates of the main and mediated effect size estimates. While we accounted for this

heterogeneity in the main effects via our moderator analyses, data limitations (i.e., small number

of effect sizes between hierarchy and the mediators) prevented us from accounting for

heterogeneity in the mediated effects (i.e., integratively testing mediation and moderation).

Indeed, no team-level meta-analysis, to our knowledge, has ever had the sample size to test a full

moderated mediation model, and other team-level meta-analysis, similar to ours, have therefore

tested moderation and mediation separately (e.g., Maynard, Mathieu, Gilson, O’Boyle, &

Cigularov, 2013; Nicolaides, LaPort, Chen, Tomassetti, Weis, Zaccarro, & Cortina, 2013;

Stajkovi, Lee, & Nyberg, 2009; Whitman, Va Rooy, & Viswesvaran, 2010). Future work will

thus benefit from larger sample sizes to detect modest effect sizes and to perform a more

integrative analysis of mediated moderation which can account for mediated effect size

heterogeneity (Aguinis et al., 2017).

Second, we note also limitations regarding sample size and, hence, statistical power.

While our cumulative sample sizes do offer considerable statistical power for main and mediated

effects, there is considerably less power for our moderating effects (although our supplementary


analyses do show most results hold when analyzing with more power). As such, we were only

able to examine moderators of the impact of hierarchy on team performance, and not on our

proposed mediators or team viability. While we expect the moderators in the literature to likely

operate upstream in our model (as shown in Figure 1, where we highlight that the literature

suggests the moderators to moderate the upstream link between hierarchy and team processes,

rather than the downstream link between processes and outcomes; see for example Bunderson et

al., 2016), future research which can have the sample size to fully test moderated mediation

models will be important. Additionally, the number of effect size estimates for certain

moderators was rather small (e.g., the number of ayclicity studies), as was the number of effect

size estimates for the relationships between hierarchy and the two mediators, making those

results susceptible to second-order sampling error (Schmidt & Hunter, 2015). Future work may

thus benefit from exploring these less represented moderators (e.g., acyclicity), mediators (e.g.,

conflict, coordination), and dependent variables in the hierarchy literature (e.g., viability).

Third, we noted when conducting this meta-analysis that a considerable portion of the

primary studies (k = 30) relied on field data. While we do not find the conclusions of these

studies to differ from those conducted in experimental settings in our supplementary analysis, we

would strongly encourage future research on hierarchy to focus more strongly on causal

methods. Relatedly, while the literature to this point has largely focused on the effects of

hierarchy on team outcomes (e.g., Bunderson et al., 2016), team outcomes can in turn affect team

inputs. We therefore strongly encourage work that not only examines the causal impact of

hierarchy on team outcomes, but also examines the causal impact of team outcomes on the

formation and stability of team hierarchies and recursive models of hierarchy.

Finally, we intentionally adopted a very broad definition of hierarchy here – of vertical


differences in socially valued resources, yet as we note in this definition and explore with our

construct-related moderators, there is much left to be understood about the different ways

hierarchies are constructed in teams in terms of the types of resources on which hierarchies are

based as well as the form or structure of vertical differences. In terms of the types of resources,

we focused here on resources most likely to be socially valued in small work teams – directly

task-related resources (Berger et al., 1972; Bunderson, 2003), but other resources, including race

or gender, may also potentially affect work team interactions, and would benefit from future

research. Additionally, within work-related differences, we did not find many significant

differences between most types of resources included in our analyses (results available from

second author upon request), but others have found resources such as power and status to differ

(e.g., Hays & Bendersky, 2015). Thus, future research would benefit from deeper understanding

of the resources underlying hierarchy. In terms of the form of hierarchy, we did not find form to

matter here, but we only looked at one conceptualization of form (Bunderson et al., 2016), and

one key form in that typology (acylicity) was not well-represented in the literature yet. Future

research could benefit from better understanding acylicity, as well as the other potential different

forms of vertical differences, such as the impact of the number or breadth of layers in a

hierarchy. Such studies could also perhaps provide more insight into when and how the

functional perspective on hierarchy may still apply.

Conclusion

In sum, our meta-analytic investigation of why and when hierarchy impacts team

effectiveness helps parse the widespread literature that has arisen on the topic of hierarchy and

identifies priorities for future research. Our study challenges several prior key propositions in the

field, including the functionalist perspective on hierarchy’s effectiveness. Our study in contrast


extends the conflict perspective on hierarchy, as we show that the relationship between hierarchy

and team effectiveness is negative on net, that this effect is mediated by conflict-enabling states,

and exacerbated in teams seemingly susceptible to conflict (i.e. teams with temporal instability,

skill differentiation, and/or mutable hierarchies). Our study therefore qualifies the functional

perspective on hierarchy and shows the need to further unpack and elaborate on the conflict

perspective on hierarchy.


References* *Papers in the meta-analytic dataset are marked with a single asterisk. Prior team meta-analyses included in our effect size benchmark (seen in Appendix B) are marked with double asterisks. Aguinis, H., Edwards, J. R., & Bradley, K. J. (2017). Improving our understanding of

moderation and mediation in strategic management research. Organizational Research

Methods, in press. doi: 10.1177/1094428115627498

Anderson, C., & Brown, C. (2010). The functions and dysfunctions of hierarchy. Research in

Organizational Behavior, 30,55-89. doi:10.1016/j.riob.2010.08.002

Anderson, C., Srivastava, S., Beer, J. S., Spataro, S. E., & Chatman, J. A. (2006). Knowing your

place: Self-perceptions of status in face-to-face groups. Journal of Personality and Social

Psychology, 91(6), 1094–1110. doi:10.1037/0022-3514.91.6.1094

Anderson, C., & Willer, R. (2014). Do status hierarchies benefit groups? A bounded functionalist

account of status. In J. T. Cheng, J. L. Tracy & C. Anderson (Eds.) The psychology of social

status (pp. 47-70). Springer: New York, NY.

Arrow H., McGrath J. E., & Berdahl J. L. (2000). Small groups as complex systems. Thousand

Oaks, CA: Sage.

Balkundi, P., & Harrison, D. A. (2006). Ties, leaders, and time in teams: Strong inference about

the effects of network structure on team viability and performance. Academy of Management

Journal, 49(1), 49-68. doi:10.5465/AMJ.2006.20785500

Banks, G. C., McCauley, K. D., Gardner, W. L., & Guler, C. E. (2016). A meta-analytic review

of authentic and transformational leadership: A test for redundancy. The Leadership

Quarterly, 27(4), 634-652. doi: 10.1016/j.leaqua.2016.02.006

Barton, A. H. (1980). A diagnosis of bureaucratic maladies. In C. H. Weiss and A. H. Barton

(eds.), Making Bureaucracies Work (pp. 27-36). Beverly Hills, CA: Sage.

Beal, D. J., Cohen, C. C., Burke, M. J., & McLendon, C. L. (2003). Cohesion and performance


in groups: A meta-analytic clarification of construct relations. Journal of Applied

Psychology, 88(6), 989-1004. doi:10.1037/0021-9010.88.6.989

*Becker, S. W., & Baloff, N. (1969). Organization structure and complex problem solving.

Administrative Science Quarterly, 14(2), 260–271. doi:10.2307/2391104

Bell, S. T. (2007). Deep-level composition variables as predictors of team performance: a meta-

analysis. Journal of Applied Psychology, 92(3), 595-615. doi: 10.1037/0021-9010.92.3.595

Bell, S. T., Villado, A. J., Lukasik, M. A., Belau, L., & Briggs, A. L. (2010). Getting specific

about demographic diversity variable and team performance relationships: A meta-analysis.

Journal of Management,37(3), 709-743. doi: 10.1177/0149206310365001

Bendersky, C., & Hays, N. A. (2012). Status conflict in groups. Organization Science, 23(2),

323-340. doi: 10.1287/orsc.1110.0734

Berger, J., Cohen, B. P., & Zelditch Jr, M. (1972). Status characteristics and social interaction.

American Sociological Review, 37, 241-255. doi: 10.2307/2093465

Bergh, D. D., Aguinis, H., Heavey, C., Ketchen, D. J., Boyd, B. K., Su, P., ... & Joo, H. (2014).

Using meta‐analytic structural equation modeling to advance strategic management research:

Guidelines and an empirical illustration via the strategic leadership‐performance

relationship. Strategic Management Journal, 37(3), 477-497.

Biggart, N. W., & Hamilton, G. G. (1984). The power of obedience. Administrative Science

Quarterly, 29(4), 540-549. doi: 10.2307/2392938

Bloom, M. (1999). The performance effects of pay dispersion on individuals and organizations.

Academy of Management Journal, 42(1), 25–40. doi: 10.2307/256872

*Bona, O. (2007). Managing team effectiveness: The impact of diversity and the mediation of

team network diversity. Tilburg University, Master thesis.


*Boone, C., & Hendriks, W. (2009). Top management team diversity and firm performance:

Moderators of functional-background and locus-of-control diversity. Management Science,

55(2), 165-180. doi:10.1287/mnsc.1080.0899

Borenstein, M., Hedges, L., Higgins, J., & Rothstein, H. R. (2014). Comprehensive meta-

analysis (Version 3) [Computer software]. Englewood, NJ: Biostat.

Bosco, F. A., Aguinis, H., Singh, K., Field, J. G., & Pierce, C. A. (2015). Correlational effect

size benchmarks. Journal of Applied Psychology, 100, 431-449.

*Brett, J. F., Pinkley, R. L., & Jacofsky, E. F. (1996). Alternatives to having a BATNA in dyadic

negotiation: The influence of goals, self-efficacy, and alternatives on negotiated outcomes.

International Journal of Conflict Management, 7(2), 121-138. doi:10.1108/eb022778

*Bunderson, J. S. (2003). Team member functional background and involvement in management

teams: Direct effects and the moderating role of power centralization. Academy of

Management Journal, 46(4), 458–474. doi: 10.2307/30040638

Bunderson, J.S., & Reagans, R.E. (2011). Power, status, and learning in organizations.

Organization Science, 22(5), 323-340. doi:10.1287/orsc.1100.0590

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. doi:10.2307/3069319

*Bunderson, J. S., Van der Vegt, G., Cantimur, Y., & Rink, F. (2016). Different views of

hierarchy and why they matter: Hierarchy as inequality or as cascading influence. Academy

of Management Journal, in press. doi: 10.5465/amj.2014.0601

Campbell, D. J. (1988). Task complexity: A review and analysis. Academy of Management

Review, 13(1), 40-52. doi: 10.5465/AMR.1988.4306775


*Cantimur, Y, Rink. F., & Van der Vegt, G.S. (2015a). Expertise and dominance pave the way

for influence attainment in the work place: The moderating role of intra-team competition. In

In. Y. Cantimur, The bright side of hierarchies: The origins and consequences of social

stratification in organisations (pp. 21-53). Groningen: University of Groningen.

*Cantimur, Y, Rink. F., & Van der Vegt, G.S. (2015b). When and why hierarchy steepness is

related to team performance. In In. Y. Cantimur, The bright side of hierarchies: The origins

and consequences of social stratification in organisations (pp. 55-82). Groningen: University

of Groningen.

*Cao, J. Top management team power inequality: The moderating roles of environmental

dynamics. Working paper.

Chan, M. E., & Arvey, R. D. (2012). Meta-analysis and the development of

knowledge. Perspectives on Psychological Science, 7(1), 79-92. doi:

10.1177/1745691611429355

Chun, J. S., & Choi, J. N. (2014). Members’ needs, intragroup conflict, and group

performance. Journal of Applied Psychology, 99(3), 437-450. doi: 10.1037/a0036363

Cohen, J. (1988). Statistical power analysis for the behavioral sciences. Hillsdale, NJ: Lawrence

Erblaum.

Cohen, J., & Cohen, P. (1983). Applied multiple regression/correlation analysis for the

behavioral sciences (2nd ed.). Hillsdale, NJ: Erlbaum.

Cohen S. G., & Bailey D. E. (1997). What makes teams work: Group effectiveness research from

the shop floor to the executive suite. Journal of Management,23(3), 239–290.

doi:10.1177/014920639702300303

Colquitt, J. A., Scott, B. A., Rodell, J. B., Long, D. M., Zapata, C. P., Conlon, D. E., & Wesson,


M. J. (2013). Justice at the millennium, a decade later: A meta-analytic test of social

exchange and affect-based perspectives. Journal of Applied Psychology, 98(2), 199-236.

doi:10.1037/a0031757

Cooper, H. (2010). Applied social research methods series: Vol. 2. Research synthesis and meta-

analysis: A step-by-step approach. Thousand Oaks, CA: Sage Publications.

**Courtright, S. H., Thurgood, G. R., Stewart, G. L., & Pierotti, A. J. (2015). Structural

interdependence in teams: An integrative framework and meta-analysis. Journal of Applied

Psychology, 100(6), 1825-1846. doi: 10.1037/apl0000027

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. doi:

10.2307/20159333

Cummings, G. (2012). Understanding the new statistics: Effect sizes, confidence intervals, and

meta-analysis. New York, NY: Routledge.

*Curşeu, P. L., & Sari, K. (2015). The effects of gender variety and power disparity on group

cognitive complexity in collaborative learning groups. Interactive Learning Environments,

23(4), 425-436. doi: 10.1080/10494820.2013.788029

DeBrock, L., Hendricks, W., & Koenker, R. (2004). Pay and performance: The impact of salary

distribution on firm-level outcomes in baseball. Journal of Sports Economics, 5(3), 243–261.

doi:10.1177/1527002503259337

DeChurch, L. A., Mesmer-Magnus, J. R., & Doty, D. (2013). Moving beyond relationship and

task conflict: toward a process-state perspective. Journal of Applied Psychology, 98(4), 559-

578. doi: 10.1037/a0032896

De Dreu, C. K. W., & Gelfand, M. J. (2008). Conflict in the workplace: Sources, functions, and


dynamics across multiple levels of analysis. In C. K. W. De Dreu & M. J. Gelfand (Eds.),

The psychology of conflict and conflict management in organizations (3–54). New York, NY:

Erlbaum.

De Dreu, C. K., & Weingart, L. R. (2003). Task versus relationship conflict, team performance,

and team member satisfaction: a meta-analysis. Journal of Applied Psychology, 88(4), 741-

749. doi.org/10.1037/0021-9010.88.4.741

De Jong, B. A., Dirks, K. T., & Gillespie, N. (2016). Trust and team performance: A meta-

analysis of main effects, moderators, and covariates. Journal of Applied Psychology, 101(8),

1134-1150. doi: 10.1037/apl0000110

De Jong, B. A., & Elfring, T. (2010). How does trust affect the performance of ongoing teams?

The mediating role of reflexivity, monitoring, and effort. Academy of Management Journal,

53(3), 535-549. doi: 10.5465/AMJ.2010.51468649

de Kwaadsteniet, E. W., van Dijk, E., Wit, A., De Cremer, D., & de Rooij, M. (2007). Justifying

decisions in social dilemmas: Justification pressures and tacit coordination under

environmental uncertainty. Personality and Social Psychology Bulletin, 33(12), 1648-1660.

doi: 10.1177/0146167207307490

De Wit, F. R. C., Greer, L.L., & Jehn, K.A. (2012). The paradox of intragroup conflict: A meta-

analysis. Journal of Applied Psychology, 97(2), 360-390. doi:10.1037/a0024844

Devine, D. J., & Philips, J. L. (2001). Do smarter teams do better a meta-analysis of cognitive

ability and team performance. Small Group Research, 32(5), 507-532. doi:

10.1177/104649640103200501

Dornbusch, S. M., & Scott, W. R. (1975). Evaluation and the exercise of authority: A theory of

control applied to diverse organizations. San Francisco, CA: Jossey-Bass.


*Doornenbal, B. M., de Jong, B. A., Nederveen Pieterse, A., & Jansen, P. G. W. Status

hierarchies in teams: The interaction between hierarchy level and hierarchy shape. Erasmus

University Rotterdam, working paper.

Dougherty, D. (1992). Interpretive barriers to successful product innovation in large firms.

Organization Science, 3(2), 179-202. doi: 10.1287/orsc.3.2.179

Duval, S., & Tweedie, R. (2000). Trim and fill: a simple funnel‐plot–based method of testing

and adjusting for publication bias in meta‐analysis. Biometrics, 56(2), 455-463. doi:

10.1111/j.0006-341X.2000.00455.x

Edmondson, A. C. (2002). The local and variegated nature of learning in organizations: A group-

level perspective. Organization Science, 13(2), 128-146. doi:10.1287/orsc.13.2.128.530

Eisenhardt, K. M., & Bourgeois, L. J. (1988). Politics of strategic decision making in high-

velocity environments: Toward a midrange theory. Academy of Management Journal, 31(4),

737-770. doi: 10.2307/256337

Ellis, P. D. (2010). The essential guide to effect sizes: Statistical power, meta-analysis, and the

interpretation of research results. Cambridge University Press: New York, NY.

*Foo, M., Sin, H., Yiong, L. (2006). Effects of team inputs and intrateam processes on

perceptions of team viability and member satisfaction in nascent ventures. Strategic

Management, 27(4), 389-399. doi: 10.1002/smj.514

*Gajendran, R. S. (2009). Leveraging diversity and technology for team performance: The role

of variety, disparity, virtuality and knowledge sharing. (Unpublished Doctoral dissertation).

The Pennsylvania State University, Pennsylvania.

Galinsky, A.D., Chou, E.Y., Halevy, N., & Van Kleef, G.A. (2012). The far-reaching effects of

power: At the individual, dyadic, and group levels. In M.A. Neale & E.A. Mannix (Eds.)


Research on Managing Groups and Teams, Vol. 15, (pp. 81-113). London, UK: Emerald

Insight.

Geyskens, I., Krishnan, R., Steenkamp, J. B. E., & Cunha, P. V. (2009). A review and evaluation

of meta-analysis practices in management research. Journal of Management, 35(2), 393-419.

doi: 10.1177/0149206308328501

*Giebels, E., De Dreu, C. K. W., & Van de Vliert, E. (2000). Interdependence in negotiation:

Effects of exit options and social motive on distributive and integrative negotiation.

European Journal of Social Psychology, 30(2), 255-272. doi: 10.1037/0021-9010.83.3.408

Gladstein, D. L. (1984). Groups in context: A model of task group effectiveness. Administrative

Science Quarterly, 29(4), 499–517. doi: 10.2307/2392936

Greer, L.L., Caruso, H.M., & Jehn, K.A. (2011). The bigger they are, the harder they fall:

Linking team power, conflict, congruence, and team performance. Organizational Behavior

and Human Decision Processes, 116, 116-128. doi: 10.1016/j.obhdp.2011.03.005

Greer, L. L., & Dannals, J. E. (2017). Conflict in teams. In R. Rico, E. Salas, & N. Ashkanasy,

The Wiley Blackwell Handbook of Team Dynamics, Teamwork, and Collaborative Working

(pp. 317-344). Somerset, NY: Wiley Blackwell.

*Greer, L. L., Dannals, J. E. & Rao, H. (2016). From cog to unique contributor: How to extract

value from hierarchy. Working Paper.

Greer, L.L., Van Bunderen, L., & Yu, S. (2017). The dysfunctions of power in teams: A review

and emergent conflict perspective. Research in Organizational Behavior, forthcoming. doi:

10.1016/j.riob.2017.10.005

*Greer, L.L., & Van Kleef, G.A. (2010). Equality versus differentiation: The effects of power

dispersion on social interaction. Journal of Applied Psychology, 95(6), 1032-1044.


doi:10.1037/a0020373

Gruenfeld, D. H., & Tiedens, L. Z. (2010). Organizational preferences and their consequences. In

Fiske S.T., Gilbert D.T., Lindzey G. (Eds.), Handbook of social psychology (5th ed., Vol. 2,

pp. 1252–1287). Hoboken, NJ: John Wiley & Sons.

Hackman, J. R. (1987). The design of work teams. In J. Lorsch (Ed.), Handbook of

organizational behavior, (315-342). Englewood Cliffs, NJ: Prentice-Hall.

*Haleblian, J., & Finkelstein, S. (1993). Top management team size, ceo dominance, and firm

performance - the moderating roles of environmental turbulence and discretion. Academy of

Management Journal, 36(4), 844-863. doi: 10.2307/256761

Halevy, N., Chou, E.Y., & Galinsky, A.D. (2011). A functional model of hierarchy: Why, how,

and when vertical differentiation enhances group performance. Organizational Psychology

Review, 1(1), 32-52. doi:10.1177/2041386610380991

Halevy, N., Chou, E. Y., Galinsky, A. D., & Murnighan, J. K. (2012). When hierarchy wins

evidence from the national basketball association. Social Psychological and Personality

Science, 3(4), 398-406. doi: 10.1177/1948550611424225

*Hambrick, D. C., Humphrey, S. E., & Gupta, A. (2015). Structural interdependence within top

management teams: A key moderator of upper echelons predictions. Strategic Management

Journal, 36(3), 449-461. doi: 10.1002/smj.2230

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.

doi: 10.5465/AMR.2007.26586096

*Haslam, A. S., McGarty, C., Brown, P. M., Eggins, R. A., Morrison, B. E., & Reynolds, K. J.

(1998). Inspecting the emperor’s clothes: Evidence that random selection of leaders can


enhance group performance. Group Dynamics: Theory, Research, and Practice, 2(3), 168–

184. doi:10.1037/1089-2699.2.3.168

Hayes, A. F., & Krippendorff, K. (2007). Answering the call for a standard reliability measure

for coding data. Communication Methods and Measures, 1, 77–89. doi:

10.1080/19312450709336664

Hays, N. A., & Bendersky, C. (2015). Not all inequality is created equal: The effects of status

versus power hierarchies on competition for upward mobility. Journal of Personality and

Social Psychology, 108(6), 867-882. doi: 10.1037/pspi0000017

*He, J., & Huang, Z. (2011). Board informal hierarchy and firm financial performance:

Exploring a tactic structure guiding boardroom interactions. Academy of Management

Journal, 54(6), 1119-1139. doi: 10.5465/amj.2009.0824

Hedges, L. V., & Olkin, I. (1985). Statistical methods for meta-analysis. San Diego, CA:

Academic Press.

Higgins, J. P. T., & Thompson, S. G. (2002). Quantifying heterogeneity in a meta-analysis.

Statistics in Medicine, 21, 1539–1558. http://dx.doi.org/10.1002/sim.1186

Hollenbeck, J. R., Beersma, B., & Schouten, M. E. (2012). Beyond team types and taxonomies:

A dimensional scaling conceptualization for team description. Academy of Management

Review, 37(1), 82-106. doi:10.5465/armr.2010.0181

Horwitz, S. K., & Horwitz, I. B. (2007). The effects of team diversity on team outcomes: A

meta-analytic review of team demography. Journal of Management, 33(6), 987-1015. doi:

10.1177/0149206307308587

Huedo-Medina, T. B., Sánchez-Meca, J., Marin-Martinez, F., & Botella, J. (2006). Assessing

heterogeneity in meta-analysis: Q statistic or I² index? . Psychological Methods, 11(2), 193-


206. doi.org/10.1037/1082-989X.11.2.193

*Hulpia, H., Devos, G., Rosseel, Y., & Vlerick, P. (2012). Dimensions of distributed leadership

and the impact on teachers' organizational commitment: a study in secondary

education. Journal of Applied Social Psychology, 42(7), 1745-1784. doi: 10.1111/j.1559-

1816.2012.00917.x

Hülsheger, U. R., Anderson, N., & 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. doi.org/10.1037/a0015978

Jehn, K. A. (1995). A multimethod examination of the benefits and detriments of intragroup

conflict. Administrative Science Quarterly, 40(2), 256 –282. doi: 10.2307/2393638

Jöreskog, K., & Sörbom, D. (1996). LISREL 8: User’s reference guide. Lincolnwood, IL:

Scientific Software.

Joshi, A., Roh, H. (2009). The role of context in work team diversity research: A meta-analytic

review. Academy of Management Journal, 52(3), 599-627. doi:

10.5465/AMJ.2009.41331491

Kahn, R.L., Wolfe, D.M., Quinn, R.P., Snoek, J.D., & Rosenthal, R.A. (1964). Occupational

stress: Studies in role conflict and ambiguity. New York: Wiley.

Keltner, D., Van Kleef, G. A., Chen, S., & Kraus, M. W. (2008). A reciprocal influence model of

social power: Emerging principles and lines of inquiry. Advances in Experimental Social

Psychology, 40, 151-192. doi: 10.1016/S0065-2601(07)00003-2

Kepes, S., Banks, G. C., McDaniel, M., & Whetzel, D. L. (2012). Publication bias in the

organizational sciences. Organizational Research Methods, 15(4), 624-662. doi:

10.1177/1094428112452760


Kilduff, G. J., Willer, R., & Anderson, C. (2016). Hierarchy and its discontents: Status

disagreement leads to withdrawal of contribution and lower group

performance. Organization Science, 27(2), 373-390. doi: 10.1287/orsc.2016.1058

**Knight, A. P., & Eisenkraft, N. (2015). Positive is usually good, negative is not always bad:

The effects of group affect on social integration and task performance. Journal of Applied

Psychology, 100(4), 1214-1127. doi: 10.1037/apl0000006

Kozlowski, S. W., & Ilgen, D. R. (2006). Enhancing the effectiveness of work groups and

teams. Psychological Science in the Public Interest, 7(3), 77-124. doi: 10.1111/j.1529-

1006.2006.00030.x

Krackhardt, D. (1994). Graph theoretical dimensions of informal organizations. Computational

Organization Theory, 89(112), 123-140.

Leavitt, K., Mitchell, T. R., & Peterson, J. (2010). Theory pruning: Strategies to reduce our

dense theoretical landscape. Organizational Research Methods, 13(4), 644-667. doi:

10.1177/1094428109345156

Lee, S. M., Koopman, J., Hollenbeck, J. R., Wang, L. C., Lanaj, K. (2015). The team descriptive

index (TDI): A multidimensional scaling approach for team description. Academy of

Management Discoveries,1(1), 88-113. doi: 10.5465/amd.2013.0001

*Lemoine, G. J., Koseoglu, G., & Blum, T. C. (2015) An integrative network approach to shared

leadership: An illustrative case with creative tasks. Working paper.

LePine, J. A., Piccolo, R. F., Jackson, C. L., Mathieu, J. E., & Saul, J. R. (2008). A meta-analysis

of teamwork processes: Tests of a multidimensional model and relationships with team

effectiveness criteria. Personnel Psychology, 61(2), 273-307. 10.1111/j.1744-

6570.2008.00114.x


LePine, J.A., & Wilcox-King, A. (2010). Editors’ comments: Developing novel theoretical

insight from reviews of existing theory and research. Academy of Management Journal, 35,

506-509.

Lipsey, M. W., & Wilson, D. (2001). Practical meta-analysis. Thousand Oaks, CA: Sage

Publications.

Magee, J.C., & Galinsky, A.D. (2008). Social hierarchy: The self-reinforcing nature of power

and status. Academy of Management Annals, 2(1), 351-398.

doi:10.1080/19416520802211628

Maner, J.K., & Mead, N.L. (2010). The essential tension between leadership and power: When

leaders sacrifice group goals for the sake of self-interest. Journal of Personality and Social

Psychology, 99(3), 482-497. doi:10.1037/a0018559

*Mannix, E. A. (1993). Organizations as resource dilemmas: The effects of power balance on

coalition formation in small groups. Organizational Behavior and Human Decision

Processes, 55(1), 1-22. doi:10.1006/obhd.1993.1021

*Mannix, E. A., & Neale, M. A. (1993). Power imbalance and the pattern of exchange in dyadic

negotiation. Group Decision and Negotiation, 2(2), 119-133. doi:10.1007/BF01884767

*Martins, L.L., Schilpzand, M. C., Kirkman, B. L., Ivanaj, S., & Ivanaj, V. (2013). A

contingency view of the effects of cognitive diversity on team performance: The moderating

roles of team psychological safety and relationship conflict. Small Group Research, 44(2),

96-126. doi:10.1177/1046496412466921

**Maynard, M. T., Mathieu, J. E., Gilson, L. L., O’Boyle, E. H., & Cigularov, K. P. (2013).

Drivers and outcomes of team psychological empowerment a meta-analytic review and


model test. Organizational Psychology Review, 3(2), 101-137. doi:

10.1177/2041386612456868

*McCurdy, H. G., & Eber, H. W. (1953). Democratic versus authoritarian: A further

investigation of group problem-solving. Journal of Personality, 22(2), 258–269.

doi:10.1111/j.1467-6494.1953.tb01810.x

*Meeussen, L., & Van Dijk, H. (2015). The perceived value of team players: a longitudinal study

of how group identification affects status in work groups. European Journal of Work and

Organizational Psychology, 1-12. doi: 10.1080/1359432X.2015.1028378

**Mesmer-Magnus, J. R., Asencio, R., Seely, P. W., & DeChurch, L. A. (2015). How

organizational identity affects team functioning: The identity instrumentality

hypothesis. Journal of Management, online preprint. doi: 10.1177/0149206315614370

Mesmer-Magnus, J. R., & DeChurch, L. A. (2009). Information sharing and team performance: a

meta-analysis. Journal of Applied Psychology, 94(2), 535-546. doi: 10.1037/a0013773

*Mitchell, R., Boyle, B., Parker, V., Giles, M., Chiang, V., & Joyce, P. (2015). Managing

inclusiveness and diversity in teams: How leader inclusiveness affects performance through

status and team identity. Human Resource Management, 54(2), 217-239. doi:

10.1002/hrm.21658

*Nederveen Pieterse, A., Hollenbeck, J. R., van Knippenberg, D., Spitzmuller, M., Karam, E. P.,

& Sleesman, D. J. (2014). The limits of self-management teams: Goal orientation diversity.

Erasmus University Rotterdam, working paper.

Ng, T. H. W. (in press). Transformational leadership and performance outcomes: Multiple

mediation pathways. The Leadership Quarterly.


**Nicolaides, V. C., LaPort, K. A., Chen, T. R., Tomassetti, A. J., Weis, E. J., Zaccaro, S. J., &

Cortina, J. M. (2014). The shared leadership of teams: A meta-analysis of proximal, distal,

and moderating relationships. The Leadership Quarterly, 25(5), 923-942. doi:

10.1016/j.leaqua.2014.06.006

O'Boyle, E. H., Banks, G. C., Walter, S., Carter, K., & Weisenberger, K. (2015). What

moderates moderators? A meta-analysis of interactions in management research. Academy of

Management Proceedings 2015(1), 14779. doi: 10.5465/AMBPP.2015.14779abstract

Pandey, S. K., & Rainey, H. G. (2006). Public managers' perceptions of organizational goal

ambiguity: Analyzing alternative models. International Public Management Journal, 9(2),

85-112.

*Patel, P., & Cooper, D. (2014). Structural power equality between family and non-family TMT

members and the performance of family firms. Academy of Management Journal,57(6),

1624-1649. doi:10.5465/amj.2012.0681

Paterson, T. A., Harms, P. D., Steel, P., & Crede, M. (2016). An assessment of the magnitude of

effect sizes: Evidence from 30 years of meta-analysis in management. Journal of Leadership

& Organizational Studies, 23(1), 66-81.

*Perry, J. L. (2014). Power and status in groups. (Unpublished doctoral dissertation). Rutgers,

The State University of New Jersey, Newark, NJ.

*Pinkley, R. L., Neale, M. A., & Bennett, R. J. (1994). The impact of alternatives to settlement

in dyadic negotiation. Organizational Behavior and Human Decision Processes, 57(1), 97-

116. doi:10.1006/obhd.1994.1006

Rico, R., Sanchez-Manaznares, M., Gil, F., & Gibson, C. (2008). Team implicit coordination

processes: A team approach. Academy of Management Review, 33(1), 163-184. doi:


10.5465/AMR.2008.27751276

*Ren, H., Gray, B., & Harrison, D. A. (2015). Triggering faultline effects in teams: The

importance of bridging friendship ties and breaching animosity ties. Organization Science,

26(2), 309-404. doi: 10.1287/orsc.2014.0944

*Ronay, R., Greenaway, K., Anicich, E.M., & Galinsky, A.D. (2012). The path to glory is paved

with hierarchy: When hierarchical differentiation increases group effectiveness.

Psychological Science, 23(6), 669–677.doi: 10.1177/0956797611433876

*Rulke, D. L., & Galaskiewicz, J. (2000). Distribution of knowledge, group network structure,

and group performance. Management Science, 46(5), 612-625. doi:

10.1287/mnsc.46.5.612.12052

Saavedra, R., Earley, P., & Van Dyne, L. (1993). Complex interdependence in task-performing

groups. Journal of Applied Psychology, 78(1), 61–72. doi:10.1037/0021-9010.78.1.61

*Sauer, N. C., & Kauffeld, S. (2013). Meetings as networks: Applying social network analysis to

team interaction. Communication Methods and Measures, 7(1), 26–47.

doi:10.1080/19312458.2012.760729

Schlaegel, C., & Koenig, M. (2014). Determinants of entrepreneurial intent: a meta‐analytic test

and integration of competing models. Entrepreneurship Theory and Practice, 38(2), 291-332.

doi: 10.1111/etap.12087

Schmidt, F. L., & Hunter, J. E. (2015). Methods of meta-analysis. Correcting error and bias in

research findings. Thousand Oaks: Sage.

Shea, G. P., & Guzzo, R. A. (1987). Groups as human resources. In K. M. Rowland & G. R.

Ferris (Eds.), Research in Personnel and Human Resource Management, vol. 5 (pp. 323–

356).


*Small, E. E., & Rentsch, J. R. (2010). Shared leadership in teams: A matter of distribution.

Journal of Personnel Psychology, 9(4), 203-211. doi: 10.1027/1866-5888/a000017

Sobel, M. E. (1982). Asymptotic intervals for indirect effects in structural equations models. In

S. Leinhart (Ed.), Sociological methodology (pp. 290-312). San Francisco: Jossey-Bass.

*Sondak, H., & Bazerman, M. H. (1991). Power balance and the rationality of outcomes in

matching markets. Organizational Behavior and Human Decision Processes, 50(1), 1-23.

doi:10.1016/0749-5978(91)90031-N

Steel, P. D. (2007). The nature of procrastination: a meta-analytic and theoretical review of

quintessential self-regulatory failure. Psychological Bulletin, 133(1), 65-94.

doi:10.1037/0033-2909.133.1.65

Steel, P. D., & Kammeyer-Mueller, J. D. (2002). Comparing meta-analytic moderator estimation

techniques under realistic conditions. Journal of Applied Psychology, 87, 96–111.

Steiner, I. D. (1972). Group processes and group productivity. New York, NY: Academic Press.

*Tarakci, M., Groenen, P. J. F., & Greer, L. L. (2016). When does power disparity help or hurt

group performance? A reconciliation of functionalist and conflict theories of power disparity.

Journal of Applied Psychology, 101(3), 415-429. doi: 10.1037/apl0000056

*Thylefors, I. (2012). All professionals are equal but some professionals are more equal than

others? Dominance, status and efficiency in Swedish interprofessional teams. Scandinavian

Journal of Caring Sciences, 26(3), 505-512. doi:10.1111/j.1471-6712.2011.00955.x

Tiedens, L. Z., & Fragale, A. R. (2003). Power moves: complementarity in dominant and

submissive nonverbal behavior. Journal of Personality and Social Psychology, 84(3), 558-

568. doi:10.1037/0022-3514.84.3.558


Tofighi, D. & MacKinnon, D. P. (2011). RMediation: An R package for mediation analysis

confidence intervals. Behavioral Research Methods, 43(3), 692–700. doi:10.3758/s13428-

011-0076-x

Torrance, E. P. (1955). Some consequences of power differences on decision making in

permanent and temporary three-man groups. In A. P. Hare, E. F. Borgatta & R. F. Bales

(Eds.), Small groups: Studies in social interaction (pp. 179–196). New York, NY: Knopf.

*Tost, L.P., Gino, R., & Larrick, R.P. (2013). When power makes others speechless: The

negative impact of leader power on team performance. Academy of Management Journal,

56(5), 1465–1486. doi:10.5465/amj.2011.0180

Tuckman, B. W. (1965). Developmental sequence in small groups. Psychological Bulletin, 63(6),

384-399. doi:10.1037/h0022100

Valentine, M. A., & Edmondson, A. C. (2014). Team scaffolds: How mesolevel structures

enable role-based coordination in temporary groups. Organization Science, 26(2), 405-422.

doi:10.1287/orsc.2014.0947

*Van Bunderen, L., Greer, L. L., & Van Knippenberg, D. (2017). Power struggles in teams: A

resource competition account of power in teams. Academy of Management Journal,

forthcoming.

*Van der Heijden, E., Potters, J., & Sefton, M. (2009). Hierarchy and opportunism in teams.

Journal of Economic Behavior & Organization, 69(1), 39-50. doi:10.1016/j.jebo.2008.09.007

*Van der Vegt, G. S., De Jong, S. B., Bunderson, J. S., & Molleman, E. (2010). Power

asymmetry and learning in teams: The moderating role of performance feedback.

Organization Science, 21(2), 347-361. doi:10.1287/orsc.1090.0452

*Van Dijk, H., Meyer, B., & Van Engen, M. L. (2015). Reconsidering the diversity-


performance relationship: Information elaboration harms performance under an inaccurate

status hierarchy. Working paper.

*Van Dijk, H., Meyer, B., & Van Engen, M. L. (2015). The origins of status in organizations:

The power of endorsements. Working paper.

Van Dijk, H., Van Engen, M. L., & van Knippenberg, D. (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. doi: 10.1016/j.obhdp.2012.06.003

Van Vugt, M., Hogan, R., & Kaiser, R. B. (2008). Leadership, followership, and evolution:

Some lessons from the past. American Psychologist, 63(3), 182–196. doi:10.1037/0003-

066X.63.3.182

Viswesvaran, C., & Ones, D. S. (1995). Theory testing: combining psychometric meta-analysis

and structural equations modeling. Personnel Psychology, 48(4), 865–885. 10.1111/j.1744-

6570.1995.tb01784.x

Wang, D., Waldman, D. A., & Zhang, Z. (2014). A meta-analysis of shared leadership and team

effectiveness. Journal of Applied Psychology, 99(2), 181-198. DOI: 10.1037/a0034531

*Wellman, E. M. (2013). Enabling shared leadership in hierarchical groups. (Unpublished

Doctoral dissertation). The University of Michigan, Michigan, USA.

**Whitman, D. S., Van Rooy, D. L., & Viswesvaran, C. (2010). Satisfaction, citizenship

behaviors, and performance in work units: A meta‐analysis of collective construct

relations. Personnel Psychology, 63(1), 41-81. doi: 10.1111/j.1744-6570.2009.01162.x

*Wolfe, R. J., & McGinn, K. L. (2005). Perceived relative power and its influence on

negotiations. Group Decision and Negotiation, 14(1), 3-20. doi: 10.1007/s10726-005-3873-8


*Woolley, A. W., Gerbassi, M. E., Chabris, C. F., Kosslyn, S. M., & Hackman, J. R. (2008).

Bringing in the experts: How team composition and collaborative planning jointly shape

analytic effectiveness. Small Group Research, 39(3), 352-371.

doi:10.1177/1046496408317792

Wright, S. C., Taylor, D. M., & Moghaddam, F. M. (1990). Responding to membership in a

disadvantaged group: From acceptance to collective protest. Journal of Personality and

Social Psychology, 58(6), 994-1003. doi:10.1037/0022-3514.58.6.994

Young-Hyman, T. (2017). Cooperating without co-laboring: How formal organizational power

moderates cross-functional interaction in project teams. Administrative Science Quarterly,

62, 179-214.

Yu, J., Downes, P. E., Carter, K. M., & O'Boyle, E. H. (2016). The problem of effect size

heterogeneity in meta-analytic structural equation modeling. Journal of Applied Psychology,

101(10), 1457-1473. doi.org:10.1037/apl0000141

Zalesny M. D., Salas E., & Prince C. (1995). Conceptual and measurement issues in

coordination: Implications for team behavior and performance. Research in Personnel and

Human Resources Management, 13, 81–115.

Zhou, Y. M. (2013). Designing for complexity: Using divisions and hierarchy to manage

complex tasks. Organization Science, 24(2), 339-355.doi:10.1287/orsc.1120.0744

Zitek, E. M., & Tiedens, L. Z. (2012). The fluency of social hierarchy: The ease with which

hierarchical relationships are seen, remembered, and heard. Journal of Personality and Social

Psychology, 102(1), 98-115. doi:10.1037/a0025345


Table 1. Meta-Analytic Correlations and Standard Deviations Matrix

1 2 3 4 51 Hierarchy 0.17 a 0.26 a 0.15 a 0.12 a

2 Coordination -0.17 (13, 960) a 0.40 b 0.14 b 0.17 b

3 Conflict 0.17 (13, 1052) a -0.6 (8, 529) b 0.20 b 0.03 b

4 Team performance -0.08 (52, 13781) a 0.29 (30, 2012) b -0.26 (32, 2049) b 0.19 c

5 Team viability -0.11 (8, 540) a 0.34 (11, 902) b -0.32 (13, 1039) b 0.32 (21, 1512) c

Note : results below the diagonal are reported as r (k ,N ), where r = corrected correlation coefficient; k = number of independent

studies; and N = cumulative sample size. Results above the diagnonal are the standard deviations for the corrected correlations. a =

correlations and standard deviations based on our own sample and analysis; b = correlations and standard deviations taken from

LePine et al. (2008); c = correlations and standard deviations taken from Maynard et al. (2013). All meta-analytic results were generated using a random effects model.


Table 2. Results of (Fixed Effects) MASEM analysis

b SE CI 95%

-0.17 0.03 -0.23 -0.11

0.21 0.03 0.15 0.27

0.23 0.03 0.17 0.29

0.17 0.03 0.11 0.23

-0.13 0.03 -0.19 -0.08

-0.18 0.03 -0.24 -0.12

Harmonic sample size 1049

ab SE CI 95%

-0.04 0.01 -0.05 -0.02

-0.04 0.01 -0.06 -0.02

-0.02 0.01 -0.04 -0.01

-0.03 0.01 -0.05 -0.02Hierarchy ® conflict ® viab

Hierarchy ® conflict ® perf

Coordination ® performance

Conflict ® performance

Note: b = path coefficient; SE = standard error; CI = confidence interval; ab = indirect effect estimate

Mediated paths

Hierarchy ® coordination ® perf

Coordination ® viability

Conflict ® viability

Hierarchy ® coordination ® viab

Individual paths

Hierarchy ® coordination

Hierarchy ® conflict


Table 3. Results of the Meta-Regression Analysis

B SE CI 95% sr

Task interdependence -0.07 0.02 -0.12 -0.03 -0.38Task ambiguity 0.14 0.04 0.07 0.21 0.45Task complexity 0.03 0.03 -0.03 0.08 0.11Membership instability -0.05 0.02 -0.10 -0.00 -0.25Skill differentiation -0.09 0.03 -0.15 -0.03 -0.36Hierarchy mutability -0.11 0.04 -0.19 -0.04 -0.35Hierarchy form: acyclicity -0.48 0.18 -0.83 -0.12 -0.30Hierarchy form: mixed 0.01 0.10 -0.18 0.20 0.01

Intercept 0.48 0.20 0.09 0.87 0.28

R 2 0.43

Moderators

Note: k = 52. B = unstandardized regression coefficient; SE = standard error; CI = confidence interval; sr = semi-partial correlation coefficient; R 2 = variance explained by the model; reference category for hierarchy form: inequality.

Table 4. Overview of Different Measurements of Intragroup Hierarchy

Hierarchy Measure k r CI 95% Typical Measurement Example Articles

Expertise Hierarchy 3 0.07 -0.11 0.24 Measuring the dispersion of expertise depth within a team, such as via the coefficient of variation or standard deviation of member expertise levels in the team

Huang & Cummings (2011), Trzebiatowski & Trevor (2016), Woolley et al. (2008)

Influence Hierarchy 15 -0.09 -0.17 0.02 Measuring the dispersion of influence in the team, such as by round robin measures of influence and then taking the coefficient of variation of member influence, or also as measured by variation in speaking time

Bunderson et al. (2016), Cummings & Cross (2003), Sauer & Kauffield (2013)

Negotiation BATNAS 7 -0.10 -0.22 0.01 Manipulating the alternatives to agreement individuals have in team negotiations, such that one person has a better alternative than the other

Giebels et al. (2000), Mannix (1993), Wolfe & McGinn (2005)

Perceived Hierarchy 4 -0.19 -0.62 0.33 Manipulating the sense of power of a leader or measuring the perceptual hierarchical climate in a team

Curseu & Sau (2013), Perry (2016); Tost et al. (2013)

Positional Hierarchy 17 -0.09 -0.16 -0.01 Measuring or manipulating the presence of different formal ranks within a team, such as the number of corporate levels in the team or manipulating the presence of a formal leader or not in a team

Greer & Van Kleef (2010); Hambrick et al. (2015), Van Bunderen et al. (2017)

Status Hierarchy 6 -0.09 -0.22 0.04 Measuring the dispersion of status in the team, such as by round robin measures of status and then taking the coefficient of variation of member status, or by taking the gini coefficient of member status markers such as board memberships

Greer et al. (2016), He & Huang (2011), Thylefors (2004)

Note: k = number of independent studies; r = corrected correlation coefficient; CI = confidence interval.


Figure 1. Conceptual model of suggested integration of different theoretical perspectives on why and when hierarchy impacts team effectiveness.


Appendix A. Overview of sample size, reliabilities, and moderators for all studies included in meta-analysis

Hierarchy Coordination Conflict Performance Viability

Article n rxx ryy rxy ryy rxy ryy rxy ryy rxy Team Inter.

Task Ambi.

Task Compl.

Member Instability

Skill Diff.

Hierarchy Mutability

Hierarchy Form

Becker & Baloff (1969) 36 1.00 - - - - 1.00 -0.40 - - M H M H ML H S

Bona (2007) 54 1.00 1.00 -0.21 - - 0.86 -0.29 0.83 -0.13 MH M M L M M C Boone & Hendriks (2009) 33 0.95 0.40 0.07 - - 1.00 0.32 - - M ML H L H ML S

Brett et al. (1996) 163 1.00 - - - - 1.00 0.04 - - L L ML H ML ML S

Bunderson (2003) 35 1.00 - - - - 0.95 0.02 - - H L L ML M ML C

Bunderson et al. (2016) 75 1.00 - - 0.52 0.09 0.84 -0.03 0.50 -0.09 M M M M M M Mix

Cantimur et al. (2015a) 57 1.00 - - - - 0.72 -0.37 - - M M M L M M A

Cantimur et al. (2015b) 72 1.00 - - 0.63 -0.03 0.85 -0.02 - - MH M M L M M S

Cao (2015) 5,167 1.00 - - - - 1.00 -0.04 - - M ML H ML H ML S Curşeu & Sari (2015) 87 - 1.00 -0.13 - - - - 0.83 -0.02 - - - - - - -

Doornenbal et al. (2015) 127 1.00 0.40 -0.24 - - 1.00 0.09 - - MH H ML MH L H S

Foo et al., (2006) 51 1.00 - - - - 1.00 -0.17 0.73 0.06 H H H M M H S

Gajendran (2009) 168 1.00 0.65 -0.19 - - 0.89 0.02 - - MH H MH MH M H S

Giebels et al. (2000) 71 1.00 - - 0.62 0.13 1.00 -0.23 - - L L ML H ML ML S

Greer, Dannals, & Rao (2016) 80 1.00 0.78 -0.19 - - 1.00 -0.09 - - H MH H M M H C

Greer & Van Kleef (2010) Study 1

42 1.00 - - 0.20 -0.01 0.45 -0.13 - - M ML M ML MH ML S

Greer & Van Kleef (2010) Study 2

40 1.00 - - 0.21 0.25 1.00 -0.09 - - ML L ML H ML ML S

Haleblian & Finkelstein (1993)

235 1.00 - - - - 1.00 -0.05 - - M ML H L H ML S




Task Ambi.

Task Compl.

Member Instability

Skill Diff.


Hierarchy Form

Hambrick et al. (2015) 365 1.00 - - - - 1.00 -0.03 - - M ML H ML H ML S

Haslam et al. (1998) 46 1.00 - - - - 1.00 -0.21 - - H H ML H ML H S

He & Huang (2011) 2,157 1.00 - - - - 1.00 -0.04 - - M ML H L H ML C

Hulpia et al. (2012) 46 - 1.00 -0.31 - - - - 0.91 -0.22 - - - - - - -

Lemoine et al. (2015) 141 1.00 - - - - 1.00 -0.07 - - H H ML H L H C

Mannix (1993) 36 1.00 - - 1.00 0.38 1.00 -0.54 - - L L ML H MH M S Mannix & Neale (1993) 675 1.00 - - - - 1.00 -0.16 - - L L ML H MH ML S

Martins et al. (2013) 196 1.00 - - 0.62 -0.02 1.00 0.21 - - MH H MH MH L H S

McCurdy &Eber (1952) 21 1.00 - - - - 1.00 0.02 - - H H ML H L H S

Meussen & Van Dijk (2015) 33 1.00 - - - - 1.00 -0.11 - - MH H MH MH L H C

Mitchell et al. (2015) 75 0.89 - - - - 0.94 -0.46 - - MH L H M MH L C

Nederveen Pieterse et al. (2014)

57 1.00 0.63 -0.09 - - 1.00 -0.10 - - H L ML H ML L S

Patel & Cooper (2014) 1,934 1.00 - - - - 1.00 -0.19 - - M ML H L H ML C

Perry (2014) 80 0.72 - - 0.57 0.19 0.91 -0.23 1.00 -0.24 M L MH MH MH ML S Pinkley et al. (1994) 142 1.00 - - - - 1.00 0.06 - - L L ML H ML ML S

Ren et al. (2015) 148 1.00 - - 1.00 -0.06 0.87 -0.04 - - MH ML H ML MH L C Ronay et al. (2012) 46 1.00 - - - - 1.00 0.51 - - M H ML H ML MH S

Rulke & Galaskiewicz (2000)

39 1.00 - - - - 1.00 -0.13 - - M L MH MH M L C

Sauer & Kauffeld (2013) 54 1.00 - - - - 0.81 -0.27 - - H ML L ML ML ML C

Small & Rentsch (2010) 60 0.90 - - 0.86 0.62 1.00 -0.17 - - MH M MH MH L H C




Task Ambi.

Task Compl.

Member Instability

Skill Diff.


Hierarchy Form

Sondak & Bazerman (1991)

144 1.00 - - - - 1.00 -0.09 - - L L ML H ML ML S

Tarakci et al. (2016) Study 1 75 1.00 - - - - 1.00 0.26 - - H H ML H L H S

Tarakci et al. (2016) Study 2 39 1.00 - - - - 0.39 -0.16 - - H ML H ML MH ML C

Thylefors (2012) 62 0.91 - - - - 0.99 -0.08 - - MH L MH M MH L S Tost et al. (2013) Study 1 20 1.00 - - - - 1.00 -0.58 - - M L ML H MH L S

Tost et al. (2013) Study 2 48 1.00 - - - - 1.00 -0.29 - - M M ML H MH ML S

Tost et al. (2013) Study 3 38 1.00 0.44 -0.31 - - 1.00 -0.39 - - M M ML H MH ML S

Van Bunderen et al. (2015) 85 1.00 - - 0.96 0.23 1.00 -0.09 - - ML L ML H ML L S

Van der Heijden et al. (2009) 20 1.00 - - - - 1.00 0.69 - - M H ML H L L S

Van der Vegt et al. (2010) 46 1.00 0.62 0.00 - - 0.90 0.14 - - M M M ML MH M S

Van Dijk et al. (2015) 96 1.00 1.00 -0.09 - - 1.00 0.05 - - H H ML H L H Mix

Van Dijk et al. (2015) 47 1.00 - - - - 1.00 -0.07 - - MH H ML H L H Mix

Wellman (2013) Study 1 87 1.00 0.38 -0.02 0.91 -0.04 0.81 -0.27 0.34 -0.05 H L M H M L C

Wellman (2013) Study 2 60 1.00 - - 0.85 -0.14 0.89 -0.25 0.87 -0.05 H H ML H L H S

Wolfe & McGinn (2005) 62 1.00 - - - - 1.00 0.03 - - L L ML H L ML S

Woolley et al. (2008) 41 1.00 1.00 0.15 - - 1.00 -0.13 - - M H ML H L H S

Notes: n = number of groups; rxx = reliability hierarchy measure; rxy = observed correlation between hierarchy and dependent variable; ryy = reliability dependent variable; L = Low; ML = Moderately Low; M = Moderate; MH = Moderately High; H = High; C = Centralization; S = Steepness; A = Acyclicity; Mix = Mixed

running head: hierarchy and team effectiveness meta-analysisset each other. second, we integratively...

Documents