Mediating effects of psychological safety in the relationship between team affectivity and transactive memory systems

• Author: Elliot Bendoly,Daniel G. Bachrach,Suzanne Zivnuska,Anthony C. Hood
• Date: 01 April 2016
• DOI: http://doi.org/10.1002/job.2050
• Published date: 01 April 2016

Mediating effects of psychological safety in the

relationship between team affectivity and

transactive memory systems

ANTHONY C. HOOD

1

*, DANIEL G. BACHRACH

2

, SUZANNE ZIVNUSKA

3

AND

ELLIOT BENDOLY

4

1

Collat School of Business, University of Alabama at Birmingham, Department of Management, Information Systems &

Quantitative Methods, Birmingham, Alabama, U.S.A.

2

Culverhouse College of Commerce, Department of Management and Marketing, The University of Alabama, Tuscaloosa,

Alabama, U.S.A.

3

Department of Management, College of Business, California State University, Chico, Chico, California, U.S.A.

4

Fisher College of Business, The Ohio State University, Columbus, Ohio, U.S.A.

Summary In this research, we develop a framework for understanding the emergence of transactive memory systems

(TMS) in project-based teams characterized by different levels of group level positive affectivity (PA) and

negative affectivity (NA). With a focus on enhancing understanding of the means of transmission, we test

the mediating role played by group level psychological safety (PS) in the relationship between team affectiv-

ity and TMS. From a sample of 107 software implementation project teams, in a lagged field study, we find

support for a mediated model in which high group NA, but not group PA, promotes environments psycholog-

ically unsafe for interpersonal risk-taking (low PS) and which are negatively associated with TMS. This study

extends prior research on the differential effects of PA and NA, by contributing to the limited research on

group affectivity, environmental antecedents of TMS, and the mediating role of PS for predicting group level

transactive processes and structures. Copyright © 2015 John Wiley & Sons, Ltd.

Keywords: transactive memory systems; team affectivity; group level psychological safety; TMS

antecedents; field study; lagged design

Increasing pressure on under-resourced project-based teams requires development and integration of specialized

expertise, mutual trusting relationships, and effective coordination routines. Research on transactive memory systems

(TMS), a set of structures and processes that provides teams with an effective means to manage scarce resources (Faraj

& Sproull, 2000), suggests that teams cope with resource constraints by dividing and sharing responsibilities for exper-

tise and tasks (Ren & Argote, 2011). A central theme underlyingthisresearchhasbeentheroleoflearning(e.g.,Lewis,

Lange, & Gillis, 2005). Specifically, transactive memory (TM) is believed to develop as group members learn about one

another’s expertise (Lewis, 2003; Wegner, 1987) and subsequently use this knowledge to develop expertise at the group

level that is specialized, credible, and well-coordinated (Lewis, 2003). Although previous research has highlighted a

number of learning-oriented factors that contribute to the development of TMS such as intimacy (Wegner, 1987), com-

munication frequency (Lewis, 2004), prior learning (Lewis et al., 2005), familiarity (Lewis, 2004), and social network

connections (Lee, Bachrach, & Lewis, 2014), factors known to hinder learning such as unfavorable perceptions of risk

and threat (Edmondson, 1999) have received less attention. As such, the purpose of the current study is to contribute to

TM theory by examining TMS development through the lens of interpersonal risk-taking and learning in teams.

For example, establishing the architecture central to a TMS is fraught with a number of potentially risky interper-

sonal behaviors such as admitting information deficiencies, declaring expertise, justifying or defending expertise

*Correspondence to: Anthony C. Hood, Collat School of Business, University of Alabama at Birmingham, Department of Management,

Information Systems & Quantitative Methods, 317 F Business and Engineering Complex, 1150 10th Ave South, Birmingham, Alabama 35294,

U.S.A. E-mail: ahood63@uab.edu

Copyright © 2015 John Wiley & Sons, Ltd.

Received 16 August 2014

Revised 22 July 2015, Accepted 23 July 2015

Journal of Organizational Behavior, J. Organiz. Behav. 37, 416–435 (2016)

Published online 2 September 2015 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/job.2050

Research Article

when challenged, and admitting lack of desire to accept responsibility for a particular expertise domain

(Hollingshead, 1998). Although TMS depends on willingness to engage in potentially risky behaviors (Larsen &

Augustine, 2008), factors encouraging or discouraging these behaviors have been largely overlooked in the TM

literature. In particular, sensitivity to and perceptions of interpersonal threat or risk are likely to influence acceptance

of risks associated with TMS. As such, we propose that teams’willingness to engage in these potentially risky

behaviors may be largely determined by the team’s dispositional affectivity.

Affectivity has been explicitly recognized as a critical stimulus in group environments (Hackman, 1992). In particular,

positive affectivity (PA) and negative affectivity (NA) are differentially associated with the direction, duration, and in-

tensity with which personal resources (e.g., time, attention, and energy) are invested in social contexts (Carver & Scheier,

1990). While PA reflects the tendency to be energetic, cheerful, and optimistic (Barsade & Gibson, 1998; Watson, Clark,

& Tellegen, 1988), NA increases the tendency to notice and ruminate over unfavorable inform ation regardin g ones’self

and others (Watson et al., 1988). These differences are likely to have an influence on the extent to which group environ-

ments are perceived as discouraging or encouraging (Edmondson & Lei, 2014) TMS-enhancing behaviors.

With this focus, we seek to contribute to the literature by deepening understanding of the antecedents leading to

TMS. First, although TMS may improve team performance (e.g., Austin, 2003; Lewis, 2003; Lewis, 2004; Liang,

Moreland, & Argote, 1995; Moreland, 1999; Moreland, Argote, & Krishnan, 1996, 1998; Moreland & Myaskovsky,

2000; Zhang, Hempel, Han, & Tjosvold, 2007), very little research has explored antecedents of TMS. The current

study contributes by developing a connection between team affectivity and TMS. Second, we also seek to deepen

understanding of the drivers of this association by examining the role played by psychological safety (PS) as a

mediator. Specifically, as shown in Figure 1, we propose that differences between high/low team-level dispositional

PA and NA lead to team-level perceptions of PS and TMS.

We begin with a review of our primary theoretical framework, TM theory. This is followed by hypothesis devel-

opment and a brief overview of multi-level theory (MLT). Understanding MLT is an essential precursor to our

methods section, as it provides the foundation necessary to understand the process through which our focal

constructs, which are typically treated at the individual and dyadic levels of analysis, emerge to form team-level con-

structs. We then provide methods, results, discussion, and implications.

Background and Theoretical Development

Transactive memory theory

According to TM theory, TM is present at the individual and dyadic levels of analysis but often forms a meaningful

phenomenon at the group level via compositional processes. At the individual level, TM refers to “…memory that is

Figure 1. The mediating roleof psychological safety in the relationship between positive and negativity affectivity and transactive

memory systems. Direct effects of positive affectivity and negative affectivity on TMS are illustrated with dashed lines

TEAM AFFECTIVTY AND TRANSACTIVE MEMORY 417

Copyright © 2015 John Wiley & Sons, Ltd. J. Organiz. Behav. 37, 416–435 (2016)

DOI: 10.1002/job