The limits of psychological safety: Nonlinear relationships with performance

This study, across five independent protocols, explored the unintended byproducts of high levels of Psychological Safety (PS).

All five protocols included multi-source data at two or more point in time – assessing a mixture of individual, team and organisational performance related to PS. The protocols included:

1. 473 knowledge employees where their in-role performance is assessed by supervisors

2. 301 hospital nurses and performance appraisals

3. 365 employees in a biomedical star-tup, with employee performance assessed by supervisors

4. A high-tech company and assessments of 82 units’ performance

?5. 229 retail stores performance data over 4 years

They argue that “While psychological safety climate is widely seen as having a positive relationship with work performance, there are compelling reasons as to why that may not … [and that] high levels of psychological safety climate can actually harm the performance of routine tasks” (p1).

Providing background, they note:

·????????PS is defined as “a shared belief that the team is safe for interpersonal risk taking” (Edmondson, 1999, p. 354), such as asking for help, challenging the status quo, revealing mistakes, and taking risks. If employees fail at something, they need not fear “a damaged reputation if the initiative fails or disapproval if it is seen as inappropriate or threatening” (p1)

·????????Prior work has shown the positive effects of PS on non-routine work behaviours, like voice, initiative, innovation, information sharing, creativity, citizenship behaviour, openness to new experience and learning. People must be able to “risk interpersonal harm” in the face of uncertainty

·????????But, not all tasks require novelty, learning and exploration. Rather, many jobs are highly standardised [** and routinised] and “allow employees little autonomy over how they are performed. Production and retail jobs, for example, are often based on repetitive and explicit tasks. Even the most complex jobs in professional areas, like health care and education, require many repetitive and standardized tasks” (p1)

·????????PS has been said to have an attribute of a ‘social shield’, leveraged to encourage and intensify outcomes involving uncertainty, risk-taking and initiative

·????????Highly standardised tasks are referred to in this study as ‘in-role performance’

·????????High PS may not always be beneficial. Edmondson even noted that “envisioning the possibility of a team with excessive safety, somehow lacking an edge to drive them forward, is not difficult.” (p1)

·????????While it’s “easy to see psychological safety climate as a simple, onesize- fits-all application, we have not mapped out, even in theoretical terms, the contexts that may be boundary conditions on its effectiveness” (p2)

·????????They propose that “where tasks are routine, high levels of psychological safety climate can harm in-role performance because it distracts employees from their core tasks first by focusing their attention on more novel tasks, and second by encouraging them to push boundaries in routine tasks where doing so is counterproductive” (p2)

·????????Further, high PS may cognitively distract workers from performance, by encouraging them to focus on novelty at the expense of routine and “The focus on novelty and risk-taking may direct their attention disproportionately to non-routine, complex tasks” (p3)

·????????Further, people performing routine tasks may be inclined to experiment with their standardised tasks, where the task may be better suited to standardisation

·????????Based on US job analysis data, 16 different ways of working/work were identified across 923 occupations. Innovation was identified in the top five requirements of just 20 occupations. The remaining 903 occupations have styles closer to routine approaches

·????????Thus, according to that data “most tasks in most jobs are likely to be routine” (p3)

·????????They propose that ‘collective accountability’, which is the shared experience of group members being held accountable for decisions and actions, can buffer the negative effect of high PS on routine tasks

Note: You’ll see that they used the term “psychological safety climate” (PSC) rather than the more common “psychological safety”. A minor annoyance I have is that psychological safety climate can either mean safety climate at the individual level (psychological), or the team environment of psychological safety (it’s climate). Whenever I come across PSC in studies, I make sure to read carefully on what citations they refer to, to determine if they mean safety climate or psychological safety.

Results

Key findings were that:

·????????Moderate levels of PS were associated with better in-role performance

·????????However, high levels of PS were associated with decreasing in-role performance

·????????Collective accountability was found to buffer the decreasing performance

While moderate PS over time provided higher in-role performance, high levels of PS were found to depress routine performance outcomes.

That is, increasing PS “up to a reasonable level” seems to improve performance, even on routine tasks.

However, for most common tasks, which are routine, very high PS may “distract and confuse workers about how to perform those tasks” (p15).

They suggest that the “simple advice” from these findings is that context matters.

Where taking (interpersonal) risks and high innovation and novelty is necessary to succeed, there may be fewer downsides of very high PS. But in the context of common and routine job tasks with clear performance standards, “there may be harm from pursuing high levels of psychological safety climate especially where the offsetting effects of collective accountability are weak” (p15, emphasis added).

These findings demonstrate the backfiring effect of very high PS.

Collective accountability was found to act like a supervisor in “redirecting attention back to important, routine tasks that need to be performed in standard ways and reducing the otherwise negative effects of that high levels of psychological safety climate on common routine tasks” (p15).

They argue that studying the limitations or negative effects of PS is challenging.

For one, linear modelling may not reveal the effects (this study employed quadratic modelling). Second, “the limits of psychological safety climate is understandably sparse because much energy goes into expanding and emphasizing its importance than pointing out its limitations. This is because the former fits much more neatly with our existing paradigms” (p15).

Link in comments.

Authors: Eldor, L., Hodor, M., & Cappelli, P. (2023). Organizational Behavior and Human Decision Processes, 177, 104255.