Examining the Role of Core Self Evaluations, Core Group Evaluations, and Individual and Team Referent Psychological Safety on Team Member Behaviours and Performance
Abstract
Psychological safety has been defined by Edmondson (1999) as “a shared belief that the team is safe for interpersonal risk taking” (p.354). Psychological safety has been found to predict a host of beneficial outcomes for teams and organizations. However, the items in the most used measure of psychological safety developed by Edmondson (1999) demonstrates inconsistency in terms of mixing referents at both the individual and team level. Individual referent (IR) psychological safety appears to be more likely to reflect individuals’ perceptions of psychological safety rather than encompassing the psychological safety of a team as a whole. On the other hand, team referent (TR) psychological safety appears to be an appropriate conceptualization of team psychological safety that aligns more closely with Edmondson’s (1999) definition of team psychological safety as a shared climate of the team.
I examined the validity of an IR measure and TR measure of psychological safety by modifying the items used in Edmondson’s (1999) measure. Overall, the findings suggested that although IR psychological safety and TR psychological safety are related, they demonstrate distinctiveness in their level of within-group agreement, in how they relate to other variables, and in evidence of incremental validity when the referent is aligned with the level of analysis. Study 1 and 2 both revealed that consistent across both studies, CSE appears to be an antecedent of IR psychological safety at the individual level, but not for TR psychological safety. At the group level, the findings relating to the relationship between CSE and TR psychological safety were unexpected but interesting. The findings suggested that higher CSE predicts lower TR psychological safety when accounting for the effect of CGE at the team level. CGE appears to be a robust predictor of IR and TR psychological safety at both the team and individual level.
In general, IR psychological safety appears to be a stronger predictor of individual-level team members’ behaviour (task performance, citizenship behaviour) whereas TR psychological safety appears to be a more potent predictor of behaviours at the team level (task performance, citizenship behaviour, counterproductive work behaviours). There were two team outcomes examined: perceived team performance (which was reported by team members) and graded team effectiveness (which was the final grade on a team project). For perceived team performance, TR psychological safety did not demonstrate incremental validity in perceived team performance above and beyond IR psychological safety as predicted. No relation was found between IR psychological safety and TR psychological safety and graded team effectiveness.
The roles of IR and TR psychological safety as a mediator were also examined. At the individual level, IR psychological safety plays a mediating role in the relationship between CSE and both task performance and citizenship behaviors. TR psychological safety plays a mediating role in the relationship between CGE and both task performance and citizenship behaviors at the individual level. Additionally, TR psychological safety mediates the relationship between CGE and both counterproductive behaviours and perceived team performance ratings at the team level.