Psychological Safety of School Administrators: A Dual Reality

Abstract

School administrators are often seen as the middle managers who are held accountable not only to their schools and communities but also to their employers. In order to successfully carry out their duties and responsibilities, school administrators must feel psychologically safe to speak up or speak out, ask a difficult question, voice an opinion, express a dissent, talk about a mistake, stand out for a position, or take a risk at work without fear of negative consequences. Psychologically precarious situations can compromise school administrators’ ability to think, feel, speak, and act, and ultimately impact their leadership, performance, and commitment to their work. Regrettably, school administrators’ psychological safety has been overlooked over the years. This research provides much-needed insights into their psychological safety in navigating this dual reality between schools and districts. The survey research garnered data from public school administrators in British Columbia, Canada, and explored school administrators’ perceptions of psychological safety at their schools and districts and its manifestation across different demographics. The multiple logit regression models show that school administrators felt less psychologically safe in their district than at their schools. The research evidence points to the importance of fostering a psychologically safe and healthy work environment in which school administrators feel psychologically safe to say: “The emperor has no clothes”.

Introduction

School administrators are often seen as the middle managers who occupy a central position in a hierarchical education system. Entrusted with implementing policies and mandates from senior leadership, they are held accountable to upper management. Simultaneously, they are responsible for day-to-day operations at a hub of competing, if not conflicting interests from various stakeholders. As a crucial link between the top management and their frontline staff and school communities, their role is essential for building trust, maintaining open communication, relaying operational feedback and insights back to the senior leadership, and driving organizational alignment and productivity. In recent years, the changing nature of their work, coupled with the increasing work demand on principalship, have further complicated their roles in navigating the work environment (Wang et al., 2022). In order to successfully carry out their duties and responsibilities, school administrators need to feel psychologically safe to “put forward an idea, ask a question, talk about a mistake, or stand out for a position” (Yildirim & Yenipinar, 2017, p. 167). However, their deeply interpersonal work adds more complexities and uncertainties to their intermediary role, and can put them in psychologically precarious situations. When, how, and with whom they speak often carries significant implications for themselves, personally and professionally. Consequently, school administrators may hold back from voicing their opinions and expressing dissent at work for fear of negative consequences to their self-image, work status, or career. Such psychological unsafety can impair their ability to think, feel, speak and act, ultimately harming their leadership, performance, and commitment (Wang, 2022). Despite its importance, research on psychological safety is mostly prevalent in business and organizational studies, primarily focusing on teamwork, collaboration, and interpersonal behaviour (Edmondson & Bransby, 2023; Frazier et al., 2017; Newman et al., 2017). Within the field of education, the existing limited research mainly addresses how school administrators foster psychological safety for their teachers and students (Ogbo et al., 2023; Shahid & Din, 2021), while the psychological safety of the school leaders themselves has been largely overlooked. This research investigates school administrators’ – specifically principals’ and vice-principals’ (VPs) psychological safety challenges amid intensified work conditions during the pandemic in the province of British Columbia (BC), Canada. It aims to elicit research evidence for building a psychologically safe and healthy work environment for school principals. The guiding questions are: How do school administrators perceive psychological safety? How is their psychological safety manifested at the school and district levels and across differing demographics?

Conceptualizing Psychological Safety

Psychological safety is the level of comfort in speaking up or speaking out without fear of negative consequences (Wang, 2022). The concept was initially introduced in the 1960s and was advanced in the 1990s by William Kahn’s (1990), who connected it to individual-level variables within organizational behaviour studies. Kahn described psychological safety as an individual’s ability “to act without fear of negative consequences, deem situations as trustworthy, develop and maintain interpersonal relationships, and engage in group dynamics without the worry of repercussions” (Kahn, 1990, as cited in Felice, 2023, p. 12). He identified four elements that played key roles in psychological safety and its development: interpersonal relationships, group and intergroup experiences, management/leadership styles, and organizational norms. In another word, psychological safety is conceptualized on three levels – the individual, the group, and the organizational (Felice, 2023; Ming et al., 2015), with a primary focus on interpersonal relationships at the workplace and their impact.

Amy Edmondson (Chen et al., 2015; Edmondson et al., 2016) later expanded this concept from the individual to the group level, coining the term “team psychological safety”. She defined it as a shared belief that the environment is safe for interpersonal risk-taking, such as asking for help or proposing ideas (Edmondson et al., 2016). Variations of Kahn’s and Edmondson’s articulation of psychological safety are widely referenced in the literature by other scholars and researchers (Ashauer & Macan, 2013; Baeva & Bordovskaia, 2015; Basit, 2017; Chen et al., 2015; Edmondson, 2019; Edmondson et al., 2016; Edmondson & Lei, 2014; Felice, 2023; Gerlach & Gockel, 2018; Roy, 2019; Shahid & Din, 2021; Weiner et al., 2021; Yildirim & Yenipinar, 2017; Zhu et al., 2019). Within the literature, psychological safety is consistently associated with trust (Shahid & Din, 2021; Yildirim & Yenipinar, 2017), work engagement, and psychological empowerment (Shahid & Din, 2021).

Conversely, the term “psychological unsafety” has emerged as a parallel construct within the literature. It is defined as “the degree to which employees perceive the risky work environment that hinder them to behave comfortably or to speak up what they think without fear of its possible negative consequences” (Yildirim & Yenipinar, 2017, p. 167). The state of psychological unsafety involves fear, shame, hesitation, embarrassment, timidity, interpersonal threats, lack of support or explanation, exclusion, and indifference (Yenipınar & Yıldırım, 2017; Yildirum & Yenipinar, 2018). Within work settings where teamwork and collaboration are needed, psychological unsafety is manifested in an environment where conflict is common, employees blame each other for failures or refrain from explaining or voicing opinions, and there exists low team satisfaction amongst group members (Yenipınar & Yıldırım, 2017; Yildirim & Yenipinar, 2017). Related terms, such as “psychologically insecure” (Yenipınar & Yıldırım, 2017) and “team insecurity” (Zhu et al., 2019), have also emerged alongside “psychological unsafety” (Yenipınar & Yıldırım, 2017; Yildirim & Yenipinar, 2017). In insecure work environments, team members tend to ignore or hide mistakes, withhold opinions, and avoid seeking help (Zhu et al., 2019). Within schools with insecure teams and environments, educators may stifle innovation due to a heightened perception of external risks (Zhu et al., 2019).

The Manifestation and Impacts of Psychological Safety

Psychological safety manifests through a workplace’s environment and its individuals in various ways. It promotes confidence, fosters teamwork, and positively influences how individuals work and relate to one another (Ashauer & Macan, 2013; Baeva & Bordovskaia, 2015; Edmondson & Lei, 2014; Itzchakov et al., 2022; Roy, 2019). In safe environments, team members develop shared trust and mutual respect amongst themselves, confident that they will not be embarrassed or shamed for taking interpersonal risks or voicing opinions (Ashauer & Macan, 2013). In workplaces with relatively high levels of psychosocial safety, employees tend to perceive their job demands to be more manageable, and communication systems are likely put into place to help monitor and manage risk (Garrick et al., 2014). In contrast, in those with low psychological safety, resources invested by administrators tend to be lacking in meeting the job demands placed upon employees (Garrick et al., 2014). Fostering this climate is not one person’s job but requires collective efforts, as “leaders and team members equally shape psychological safety” (Gerlach & Gockel, 2018, p. 307). Psychologically safe settings enable employees to fully share their cognitive, physical, and emotional resources (Felice, 2023).

Research also shows that power distance – the degree of inequality and acceptance of hierarchical power within an organization – can fundamentally impact psychological safety (Felice, 2023). In organizations with high power distance with centralized authority, those lower in the hierarchy tend to feel less empowered and hesitant to voice their opinions and concerns for fear of negative repercussions (Felice, 2023). This dynamic reinforces silence and compliance and hinders open communication, and is detrimental to psychological safety. Thus, the mindful exercise of power and the active reduction of social barriers between in-groups and out-groups are critical to establishing and sustaining psychological safety.

Within schools, heightened psychological safety and autonomy reduce staff conflicts, creating a less divided team and a more unified workforce (Itzchakov et al., 2022). Research in Russian secondary schools found a correlation between teachers’ psychological safety and students’ psychological well-being (Baeva & Bordovskaia, 2015). When teachers experienced high psychological safety, they exhibited greater well-being and emotional comfort, which they directed towards their students. These scholars assert that psychological safety satisfies “the needs for personal trust and communication, creates the referential importance of the environment (gives one a sense of belonging), and promotes mental health” (Baeva & Bordovskaia, 2015, p. 89).

A school’s psychological safety climate (PSC) acts as a critical job resource for school principals, moderating the impact of increased job demands on their health, engagement, and daily recovery from their work in schools (Garrick et al., 2014). When individuals are unable to adequately recover from their daily work-related fatigue, chronic fatigue and burnout can result (Garrick et al., 2014; Winwood et al., 2006). Since a school’s psychological safety climate stems from and is shaped by management (Garrick et al., 2014), investing in developing healthy, supportive, and trusting climates is paramount for education departments, human resources, and school management teams (Edmondson & Lei, 2014; Felice, 2023; Garrick et al., 2014; Itzchakov et al., 2022). Schools high in psychological safety possess “… a culture where people feel supported and pushed to engage in the interpersonally ‘risky behavior’ or learning and change, whereas those with lower levels of [psychological safety] can often be characterized as collegial and/or caring but lacking in terms of collective push to learn and change” (Weiner et al., 2021, p. 11).

Psychological Safety and School Administrators

Within educational environments, psychological safety as a condition is characterized by meeting needs for belonging, personal trust, and communication, promoting mental health, and ensuring an absence of psychological violence in social interactions (Baeva & Bordovskaia, 2015). It is a critical factor for healthy schools, and school leaders play a significant role in fostering it (Edmondson & Lei, 2014; Felice, 2023; Håkansson et al., 2021; Shahid & Din, 2021; Weiner et al., 2021). However, the existing studies on psychological safety focus almost exclusively on how principals create psychological safety for others, in particular for teachers and between teachers and students. Working in a hierarchy that is marked by a deeply interpersonal nature, school administrators are inevitably impacted not only by the psychological safety they create for others, but also by the psychological safety they experience themselves. Despite this, there is little research on school administrators’ own psychological safety.

As middle managers, school administrators are accountable to both their students and school communities and their employers (e.g., school districts as in the case of British Columbia, Canada). This dual accountability creates a hierarchically structured work environment where experiencing psychological unsafety is a distinct possibility. The levels of psychological safety among school administrators are affected not only by school factors such as the composition of the school population, school type, size, and location, but also by broader structural or organizational characteristics of the education systems. School districts often determine power, control, autonomy and responsibilities principals have (Devos et al., 2007). While school administrators often receive operational guidelines from district or state leaders regarding school operations, but they are typically excluded from decision-making processes at the district level. Many have expressed dissatisfaction with what they perceived as their districts’ limited perspective on their role (Bantwini & Moorosi, 2018). They feel that their districts view them as passive policy implementers as much as active contributors to instructional leadership. There was a lack of meaningful engagement of school administrators within the district regarding ongoing curriculum and policy reforms at the national, provincial, and district levels. This limited involvement of principals in crucial decision-making processes reflects a top-down culture that compromises their psychological safety.

This dynamic fosters a lack of trust in the senior officials among school administrators (Bantwini & Moorosi, 2018). This is reflected in their discontentment with their senior leadership due to the lack of support from various stakeholders, particularly from superintendents (De Jong et al., 2017). Instead of feeling empowered and supported to do their job, many report being micromanaged by the school boards, which keeps them in check. When psychological safety is low, principals feel their autonomy over daily school operations is constrained by the district and/or the union (Weiner et al., 2021) – a situation exacerbated in BC, where school administrators are not part of the teachers’ union. Conversely, in districts that encouraged independent decision-making and where the psychological safety at the schools was high, principals felt empowered to take actions they deemed necessary for their schools (Weiner et al., 2021).

Conceptual Framework

This research is guided by a conceptual framework that outlines the key variables and their expected relationships, drawing from existing literature. The framework was developed to serve as a roadmap for inquiry, with details presented below.

While school administrators are expected to create a school environment in which students and teachers feel psychologically safe, there is equally a strong need to ensure their own psychological safety is supported. Research shows that school administrators’ psychological safety is significantly influenced by their relationship with the senior leadership at the district level (Wang, 2022). Those with trusting relationships with their senior leadership tend to feel more empowered and have a greater sense of psychological safety at work (Wang, 2022). This study examines psychological safety at the district level and explores to what extent district level psychological safety contributes to the overall psychological safety of school administrators. Drawing on literature and previous studies, district-level psychological safety was measured through school administrators’ comfort and fear in speaking up, asking difficult questions, raising problems, and challenging unwritten rules and assumptions.

Furthermore, school administrator autonomy coupled with district professional support enhances effectiveness in their performance at work (Weiner et al., 2021). During the COVID-19 pandemic, principals reported that their district leaders supported their autonomy in decision-making within their schools and how “…they embraced decision-making once district or state leaders set a general framework for school operations” (Weiner et al., 2021, p. 12). District or educational leaders have a powerful role to create psychologically safe spaces for today’s school administrators as they navigate complex decisions and optimal actions in supporting their school and students. Thus, district psychological safety measurement also considers whether school administrators can count on their district to defend them, if they are being valued and accepted (i.e., a sense of belonging).

Similarly, school-level psychological safety was measured along risk-taking, respect, belonging, and fear and comfort, taking into consideration school-specific interactions and dynamics. At both levels, school administrators’ demographics, informed by relevant literature, were considered for analysis, such as their gender, age, race, principalship, work experience, and school type, location, and size. Given the limited research on school administrators’ psychological safety, this study is exploratory. It aims to understand the dual reality concerning school administrators’ psychological safety at school and district levels, providing a foundation for future research along a similar line.

Research Methodology

This research is part of a larger study on Subversive Leadership and The Art of War. The study employed a sequential mixed methods design (Tashakkori & Teddlie, 2010), drawing on data sources from interviews, focus groups, and an online survey, with pandemic-related adjustments. This approach leverages the strengths of both quantitative and qualitative research to mitigate limitations of both approaches, like issues with knowledge generalization, research credibility, and researcher bias (Creswell, 2014). Data for this article is primarily from the online survey that garnered data on school administrators’ perceptions of their psychological safety and its impacting factors. The instrument was developed based on previous research and refined through a literature review and interviews and focus groups conducted in collaboration with the British Columbia Principals’ and Vice-Principals’ Association (BCPVPA) (see details in Wang, 2022). The survey contained 48 questions on areas such as psychological safety, safety and well-being, work relationships, organizational leadership and management, and demographic information, and took between 45–60 minutes to complete. School administrators across British Columbia’s 60 publicly funded English and French school districts were invited via email to participate in this survey research. The final response rate after data cleaning was 28.0%.

Data Analysis

Prior to data analysis, missing values were investigated using the Expectation–maximization (EM) algorithm (Allison, 2011). The internal consistency of the survey questions was assessed using Cronbach’s Alpha, with a threshold of .60 (Taber, 2018). Principal Component Analysis (PCA) identified the underlying factors and their associated items (Ryan et al., 2019). Exploratory factor analysis (EFA) with varimax rotation reduced the dataset to a smaller set of variables, retaining loadings greater than .60 (Samuels, 2017). Correlation analysis first investigated the collinearity among the predictors before examining the linear association between the overall psychological safety and the predictors. Ordinal regression analysis using a logit function was conducted to identify significant predictors of overall psychological safety at work. The ordinal logit model estimates the impact of multiple predictors on ordered response variables when the outcome variables include more than two categories (Chen & Hughes, 2004; Hirk et al., 2020). A PseudoR² value was reported as an equivalent statistic to R² to identify how much variation in the model was explained by the predictors. All analyses were performed using R Version 4.4.1.

A total of n = 323 responses were analyzed. The descriptive statistics (n, %) for demographic characteristics are presented in Table 1. The sample largely represents the BCPVPA membership: 68.7% principals and 31.3% VPs from elementary (73.4%) and secondary schools (26.6%). About 61.3% of the sample identified as female and 38.7% identified as male. Geographically, 30.7% worked in the rural area (e.g., villages, small towns, and towns with population sizes fewer than 3,000 to up to 100,000) and 69.3% in the urban area ¹ (e.g., cities and larger cities from 100,000 to over one million people). Participants had an average of 9.8 years of work experience (SD = 5.31); the majority of them had over five years of work experience (78.0%), while 22.0% had less than five years. The most common age groups were 50–59 years old (47.4%), and 40–49 years old (41.5%), followed by 60 and above (5.9%) and 30–39 years old (5.3%). The age and work experience distribution indicates a cohort of younger administrators with less work experience. Racially, most respondents identified as white (92.6%), with a small percentage from other racial/ethnic backgrounds (7.4%). School sizes varied from 300 and smaller (34.4%), 301–600 (39.3%, and greater than 600 (26.3%).

Table 1.

Sample Characteristics

Demographics	Characteristics	N	%
Job category	0 Vice-principal	101	31.3
	1 Principal	222	68.7
Gender	0 Female	198	61.3
	1 Male	125	38.7
Age	1 30–39	17	5.3
	2 40–49	134	41.5
	3 50–59	153	47.4
	4 60 and above	19	5.9
Race	0 PIPOC	24	7.4
	1 White	299	92.6
School type	0 Elementary	237	73.4
	1 Secondary	86	26.6
Location	0 Rural	99	30.7
	1 Urban	224	69.3
Work experience	0 ≤5 years	71	22.0
	1 >5 years	252	78.0
School size	1 ≤300	111	34.4
	2 301–600	127	39.3
	3 >600	85	26.3

Research Findings

This section presents findings on school administrators’ perceptions of their overall psychological safety at work, followed by the results of reliability analysis, exploratory factor analysis (EFA), multicollinearity assessment, model selection, and the final ordinal regression model.

Perceptions of Overall Psychological Safety at Work

In general, survey results revealed mixed sentiments among school principals and VPs regarding their psychological safety at work. Approximately 22.9% of the respondents expressed that their psychological safety at work is Very Poor (5.6%) or Poor (17.3%), 26.3% of them felt Neutral about their psychological safety, and a little over half of the respondents indicated their psychological safety is Good (43.7%) or Excellent (7.1%).

School administrators’ psychological safety manifested differently at the school and district levels (see Table 2). At the school level, over half of the principals and VPs “often” or “always” felt that they were being “respected by teachers and school staff in this school” (81.8%), their “skills and talents are valued and utilized” (68.7%), they were “able to bring up problems and tough issues” (60.7%), and felt “safe to take risks in their leadership at this school” (53.2%). However, more than one-fifth of them responded that they felt they had to “be careful not to make mistakes around their staff” (32.2%) and “when they make a mistake, it is often held against them” (19.5%). At the district level, fewer administrators reported positive experiences. Most of them indicated that they “often” or “always” “feel free to discuss issues with the district staff” (42.1%), or “feel comfortable asking difficult questions” (36.2%), and “are able to bring up broblems and tough issues to senior management” (35.9%). However, less than a quarter of the respondents indicated that “they can speak up without any fear of retaliation” (22.9%), or “they feel comfortable challenging unwritten rules and assumptions” (21.7%).

Table 2.

Psychological Safety at School and District Levels

	Never	Rarely	Sometimes	Often	Always
School level
c) I feel when I make a mistake at this school, it is often held against me	14.2	39.0	27.2	10.5	9.0
b) I feel I have to be careful not to make mistakes around my staff	7.7	31.3	28.8	21.4	10.8
f) I can speak up without any fear of negative consequences at this school	5.3	13.0	36.2	36.8	8.7
a) I feel safe to take risks in my leadership at this school	2.5	6.5	37.8	40.2	13.0
g) I am able to bring up problems and tough issues in this school	2.2	6.5	30.7	44.0	16.7
j) Working with members of this school, my skills and talents are valued and utilized	1.9	7.1	22.3	52.0	16.7
d) I feel respected by teachers and school staff in this school	1.9	2.5	13.9	51.1	30.7
District level
e) I feel comfortable challenging unwritten rules and assumptions	11.8	29.4	37.2	18.3	3.4
f) I can speak up without any fear of retaliation in the school district	14.9	29.1	33.1	17.0	5.9
d) I am able to bring up problems and tough issues to senior management	7.4	19.5	37.2	27.9	8.0
c) I feel comfortable asking difficult questions	7.1	17.3	39.3	30.3	5.9
a) I feel free to discuss issues with the district staff	3.7	14.2	39.9	33.1	9.0

The numbers in bold are the highest percentages for each item.

Reliability Analysis, Exploratory Factor Analysis (EFA) and Multicollinearity

The internal consistency for the psychological safety items at the school and district levels was examined using Cronbach’s alpha. The overall Cronbach’s Alpha was high, with α = .88, Mean = 3.3, SD = .66, exceeding the threshold of .60 and suggesting acceptable reliability. The individual item Cronbach’s Alphas are reported in Table 3.

Table 3.

Descriptive Statistics and Reliability Analysis Results

School level	Mean	SD	Cronbach Alpha
a) I feel safe to take risks in my leadership at this school	3.5	.89	.87
b) I feel I have to be careful not to make mistakes around my staff	3.0	1.13	.88
c) I feel when I make a mistake at this school, it is often held against me	3.4	1.13	.87
d) I feel respected by teachers and school staff in this school	4.1	.84	.88
f) I can speak up without any fear of negative consequences at this school	3.3	.98	.87
g) I am able to bring up problems and tough issues in this school	3.7	.91	.87
j) Working with members of this school, my skills and talents are valued and utilized	3.7	.88	.88
District level
a) I feel free to discuss issues with the district staff	3.3	.95	.87
c) I feel comfortable asking difficult questions	3.1	.99	.87
d) I am able to bring up problems and tough issues to senior management	3.1	1.04	.87
e) I feel comfortable challenging unwritten rules and assumptions	2.7	1.00	.88
f) I can speak up without any fear of retaliation in the school district	2.7	1.10	.87

Principal Component Analysis (PCA) was used to identify the underlying factors and their associated items for psychological safety at work. The items with low EFA loadings (<.60) were excluded from the analysis results. The analysis results on the remaining high EFA loadings (see Table 4) suggested a two-factor structure: one at the school level, and one at the district level, each with at least three items with loadings greater than .60 (Samuels, 2017). The proportion of the total variance explained by the retained factors was about 30.1% at the school level and 59.2% at the district. The Kaiser-Meyer-Olkin measure of sampling adequacy exceeded .60, along with the significant result of Bartlett’s test of sphericity (p ≤ .001), which confirmed the factorability of study variables (Kaiser, 1974).

Table 4.

EFA Factors Loadings at School and District Levels

Items	Factor 1	Factor 2
Psychological safety at the school level
c) I feel when I make a mistake at this school, it is often held against me	0.784
g) I am able to bring up problems and tough issues in this school	0.746
f) I can speak up without any fear of negative consequences at this school	0.711
j) Working with members of this school, my skills and talents are valued and utilized	0.707
d) I feel respected by teachers and school staff in this school	0.692
b) I feel I have to be careful not to make mistakes around my staff	0.659
a) I feel safe to take risks in my leadership at this school	0.623
Psychological safety at the district level
d) I am able to bring up problems and tough issues to senior management		0.895
a) I feel free to discuss issues with the district staff		0.836
c) I feel comfortable asking difficult questions		0.817
f) I can speak up without any fear of retaliation in the school district		0.794
e) I feel comfortable challenging unwritten rules and assumptions		0.680
Total of variance explained by each factor	30.10%	59.20%

A correlation matrix was used to investigate the collinearity among predictors (see Table 5). The correlation coefficient between school and district level psychological safety is r = .42, smaller than the threshold (r < .80), indicating that there is no multicollinearity among predictors (Shrestha, 2020). The correlation matrix also shows that all predictors are significantly associated with the overall psychological safety (PS).

Table 5.

Correlation Matrix

	PS	School	District
PS	1.000	.616**	.540**
School		1.000	.419**
District			1.000

**Correlation is significant at the 0.01 level.

Model Identification and Selection

Factor analysis confirmed two distinct factors, indicating that psychological safety is manifested at two different levels among school administrators: school-level (factor 1) and district-level psychological safety (factor 2). A multiple ordinal logit regression analysis was conducted to explore which level of psychological safety would contribute to the overall psychological safety and to what extent. The ordinal logit model is suitable for ordinal outcome variables (Liu, 2009). In this study, the outcome variable consisted of ordered categories (e.g., the 5-point Likert scale), and the ordinal logit model was thus employed to identify the potential significant predictors – the school and district level psychological safety on the overall psychological safety at work (Hirk et al., 2020). Individual, professional, and organizational variables (e.g., gender, years of work experience, principalship, and school location) were also included in the model. Several regression models were specified and compared to explore the best model fit to the data using Akaike Information Criterion (AIC) (Akaike, 1974) and Bayesian Information Criterion (BIC) scores (see Table 6). In regression analysis, the AIC is a technique to estimate the likelihood of the best model for predicting future observations (Chakrabarti & Ghosh, 2011) and is a powerful investigative tool to compare models with different confounding variables (Akaike, 1974) and explore different hypotheses and identify the best model fit (Portet, 2020). BIC is another criterion (Stone, 1979) for model selection that measures the trade-off between the model fit and the complexity of the model (Spiegelhalter et al., 2002). Table 6 presents the model fit comparison and model selection criteria based on both AIC and BIC scores, where a lower AIC or BIC scores indicate a better fit to the given data.

Table 6.

Model Fit Comparison Based on AIC and BIC

Model specification	K	AIC	BIC	Residual deviance	Log- likelihood	McFadden’s R 2
Model 1	14	698.639	751.526	670.639	−335.32	24.00%
Model 2	10	691.091	728.868	671.091	−335.55	23.95%
Model 3	9	690.130	724.129	672.130	−336.07	23.83%
Model 4	8	690.294	720.515	674.294	−337.15	23.58%
Model 5	7	691.658	718.102	677.658	−338.83	23.21%

M1(Full model): Q16WB_EM ∼ SchoolPSn3+ DistricPSn3+ YrsExp + Job +Race +Age +Gender + Loctn + SchType +SchSize.

M2: Q16WB_EM ∼ SchoolPSn3+ DistricPSn3+ YrsExp + Job + Age + Gender.

M3: Q16WB_EM ∼ SchoolPSn3+ DistricPSn3+ YrsExp + Job + Gender.

M4: Q16WB_EM ∼ SchoolPSn3+ DistricPSn3+ YrsExp +Gender.

M5: Q16WB_EM ∼ SchoolPSn3+ DistricPSn3+ YrsExp.

Model 1 presents the full model that includes all predictors. The non-significant predictors with the highest p-values (p > .05) were excluded for each alternative model, and the AIC and BIC scores were reported and used to estimate model fit. The ordinal logit regression results show that school size, race, school location, and school types are the variables with the highest p-values and were sequentially excluded in the alternative models. All five alternative models explain the same Pseudo R² value of approximately 23.0% to 24.0%, indicating an excellent model fit (Pseudo R² >20.00%) (McFadden et al., 1977). Model 3 had the smallest AIC score (AIC = 690.130), suggesting the best predictive fit to the data while Model 5 had the lowest BIC score (BIC = 691.658), indicating an alternative best model fit with less model complexity for the data. The Brant test examined whether the proportional odds assumption (parallel lines assumption) is met. The non-significant Brant test for models 1 to 5 suggests that the effects of the explanatory variables are consistent across the different thresholds of the outcome variable – Overall Psychological Safety (p > .05) (Brant, 1990). Thus, the parallel lines assumption in the five alternative models was not violated.

Through model fit comparison, Model 5 with the smallest BIC score was selected as the best model fit to the given data. The ordinal logit regression results (see Table 7) showed that both school and district level psychological safety positively impact school administrators’ overall psychological safety. For every one unit increase in psychological safety at the school level, the odds of having a high level of overall psychological safety are about (OR = 10.028) 10 times greater, assuming other variables are held constant (OR = 10.028, 95% CI [6.209, 16.546], p < .001). However, at the district level, for every one unit increase in psychological safety, there will be a 2.910 unit increase in the overall psychological safety of school administrators, holding other variables constant (OR = 2.910, 95% CI [2.102, 4.065], p < .001). This discrepancy highlights a dual reality: school administrators felt psychologically safer at their schools, but less so in their district. They are three times more likely to experience psychological safety at their school than in their district. At their school, school administrators felt safer bringing up problems and issues, speaking up, taking risks, and making mistakes, and tended to feel respected and valued. However, when interacting with their district, they felt less comfortable speaking up, sharing ideas, discussing issues, and asking difficult questions.

Table 7.

Analysis Results of Ordinal Logit Model (m5)

Variables	β	SE	p - value	t	OR	95% CI for exp(β)
						Lower	Upper
School level	2.305***	.250	<.001	9.233	10.028	6.209	16.546
District level	1.068***	.168	<.001	6.359	2.910	2.102	4.065
Work experience	−.747*	.270	<.01	−2.764	.474	.277	.802

Note: OR = odds ratio; SE = standard error; CI = confidence interval.

***significant at p < .001, while *significant at p < .05. Model 5: PS (Q16) ∼ School + District + YrsExpP_VP_5; AIC = 691.658; BIC = 718.102.

Of all the demographic variables analysed – including principalship, gender, age, race, school type, location, and school size – only years of work experience was found to be a statistically significant predictor of psychological safety. The results indicate that years of work experience are negatively associated with school administrators’ overall psychological safety. Each additional year increase in work experience was associated with a 52.6% (1–.474 = .526) decrease in the odds of having a higher level of overall psychological safety (OR = .474, 95% CI [0.277, 0.802], p < .01). The results seem to be counterintuitive considering that, with increased work experience, school administrators are more likely to develop political savvy and strategies to cope with psychologically unsafe situations. Nevertheless, it is also likely that, as they are in the profession longer, they tend to be more aware of the implicit organizational rules and assumptions and the negative consequences of challenging them. Thus, the fear of negative repercussions leads to greater caution and hesitancy to speak out, express their opinions, dissent, stand for a position, or challenge the status quo at work. The remaining demographic variables were non-significant, indicating that school administrators’ psychological safety does not meaningfully differ across gender, age, racial background, school type, geographic location, or school size. These factors are not significant predictors of their psychological safety, suggesting that the experience of psychological safety may be a universal aspect of the administrative role itself, rather than being shaped by these individual or contextual demographics.

Discussion

School administrators operate within a hierarchical structure defined by complex interactions and relationships with stakeholders at various levels. Elements of this hierarchy – such as status gaps and power distance – significantly influence individuals’ perception and experience of psychological safety (Edmondson et al., 2016; Felice, 2023; Weiner et al., 2021).) Power distance reflects the degree of inequity that exists between individuals and structures, often manifested through control over resources (Appelbaum et al., 2020). Greater power distance has been associated with autocratic and hierarchical decision-making (Felice, 2023) and can leave those lower in an organization’s hierarchy feeling powerless, experiencing learned helplessness, and believing their voices and opinions carry little to no weight. This is especially evident in school administrators’ relationships with senior leadership at their school districts. Their psychological safety is markedly lower at the district level, making them less inclined to speak up, raise problems and tough issues, ask difficult questions, or discuss issues openly with the district. Edmondson et al. (2016) stated that “hierarchy, the degree of authority, and respect afforded to individuals based on their position in a social system, inhibits psychological safety…” (p. 69). Those who accept hierarchical structures at work are also more likely to obey rules, be compliant, and avoid disagreements (Felice, 2023). As middle managers, school administrators feel less comfortable to challenge counterproductive unwritten rules and assumptions, even when aware of their drawbacks (Wang, 2022).

The hierarchical structure can create in-groups and out-groups, distinguishing school administrators from their district leaders and affecting their perception and experience of psychological safety at work. District leaders, positioned at the top of their organizational hierarchy with arguably the most power, represent an out-group that can attenuate school administrators’ sense of psychological safety, keeping them in a state of self-consciousness and inauthenticity (Felice, 2023; May et al., 2004). School administrators may feel pressured to conform to organizational norms, even when doing so undermines their level of psychological safety within their work environments (Felice, 2023).

Prior research has confirmed that individuals with higher status within workplaces have higher levels of psychological safety in comparison to those with lower status (Edmondson et al., 2016). Those with more power can influence decisions such as promotions and reviews as well as other decisions that can affect the career growth of their subordinates (Basit, 2017). This reality can impact trust and perceived safety that employees have with their supervisors in their organization’s hierarchy. When leaders in an organization or workplace actively work towards reducing status gaps, it can help enhance psychological safety (Edmondson et al., 2016; Felice, 2023; Weiner et al., 2021). In fact, when power is shared within a workplace, it can further foster psychological safety (Yildirim & Yenipinar, 2017). A climate of psychological safety can mitigate the interpersonal risks inherent in organizational hierarchies (Edmondson & Lei, 2014). Supervisors and district leaders can improve principals’ psychological safety by developing trust and the sense of safety and obligation in their interactions with their subordinates through effective communication (Basit, 2017, p. 715). When employees cannot trust their supervisors or other authority figures, their sense of psychological safety at work can be compromised (Basit, 2017; Edmondson et al., 2016). In comparison, when employees view their supervisors or those in positions of authority with trust and perceive that they are capable of creating these conditions to overcome interpersonal risks, it can help foster psychological safety (Edmondson, 2019; Shahid & Din, 2021). Thus, a trusting and supportive relationship between school and district leadership can help create a positive organizational culture in which school administrators feel supported and valued with greater sense of psychological safety.

Literature suggests that men and women can vary in their experience and perception of psychological safety (Felice, 2023; Gerlach & Gockel, 2018; Yenipınar & Yıldırım, 2017). While safe environments were vital for female employees to feel that their opinions and voices matter (Felice, 2023), female employees have been reported to experience less psychological safety than their male counterparts (Felice, 2023; Gerlach & Gockel, 2018; Yenipınar & Yıldırım, 2017). However, this study found no significant difference on psychological safety between female and male school administrators, implying working in the shared professional roles, they develop similar stress responses towards psychological safety. But further research is needed to explore this finding.

Research shows that years of work experience can influence how psychologically safe or insecure an individual may feel within their workplace and with their colleagues (Gerlach & Gockel, 2018; Persson et al., 2021; Yildirim & Yenipinar, 2017). For example, compared to their later-career colleagues, novice teachers were found to have lower levels of psychological safety (Edmondson et al., 2016). As their experience and position status increased, their perceptions of psychological insecurity or unsafety decreased. Nevertheless, work experience appears to be having an opposite effect on school administrators’ psychological safety in this research. The longer school administrators are in the profession, the more likely they report a low level of psychological safety. Years of work experience does not appear to buffer against numerous adverse psychosocial factors in this context.

Principalship also plays a role in psychological safety. Previous research shows that while some vice-principals (VPs) feel psychologically unsafe with their principals at their own schools, principals’ psychological unsafety tend to occur in their interaction with the districts (Wang, 2022). Additionally, BIPOC school leaders, who are more frequently placed in VP positions in BC, often face psychological unsafety that tends to overlap with racial discrimination they are experiencing, putting them in a more psychologically precarious situation (Wang, 2022). However, this research found no statistically significant differences based on principalship and race, suggesting that school administrators share common experiences regardless of their principalship or race/ethnicity.

The fact that most of the demographic variables are non-significant predictors of psychological safety suggests that psychological safety may be a function of role-specific challenges and systemic structures – such as hierarchical relationships with district leadership – rather than individual or school-level demographics. Further research is needed to confirm these findings.

Moving Forward

Psychological safety is fostered when individuals’ input and voices are actively sought and heard (Edmondson, 2019; Edmondson et al., 2016; Gerlach & Gockel, 2018; Zhu et al., 2019). Individuals feel empowered when their voices and input are invited for contribution and consideration (Zhu et al., 2019). Leaders – including school administrators, district officials, and ministry personnel – are “obliged to set direction, to invite crucial, relevant input to clarify and improve on that direction, and to create conditions for continued learning to achieve excellence” (Edmondson, 2019, p. 17). Proactively seeking the input of others is critical for building psychological safety, and in fact, inviting engagement from others has been considered an essential activity in a leader’s toolbox (Edmondson, 2019; Edmondson et al., 2016). Genuine invitations require “adopting a mindset of situational humility and proactive inquiry” (Edmondson, 2019, p. 17). As Gerlach and Gockel (2018) note that proactive inquiry can be important as not everyone feels welcome in sharing their concerns or ideas, and explicit and repeated invitations and encouragement are often necessary.

Developing trusting relationships with supervisors and district leaders is also key in enhancing school principals’ psychological safety. Open, honest, frequent, and transparent communication between principals and educational leaders builds trusting relationships (Basit, 2017; Itzchakov et al., 2022) and reinforces psychological safety (Baeva & Bordovskaia, 2015; Edmondson et al., 2016; Edmondson & Lei, 2014; Felice, 2023; Gerlach & Gockel, 2018; Itzchakov et al., 2022; Shahid & Din, 2021; Yildirim & Yenipinar, 2017). Within schools, mutual respect and trust between school administrators and staff makes a significant contribution to psychological safety. In association with psychological safety, Carmeli et al. (2009) identified five components of high-quality relationships: mutuality, connectivity, positive regard, emotional carrying capacity, and tensility. The authors posited that high-quality relationships lead to psychological safety.

Improving school administrators’ psychological safety should begin “at the preparatory level”, equipping people with “ways of working that may help them navigate the strong headwinds that reduce psychological safety at work” (Edmondson et al., 2016, p. 80). Schools and school leaders should be educated and trained on the benefits of psychological safety, and how psychological safety is often linked to organizational outcomes (Shahid & Din, 2021). Currently, there is a plethora of research and discussion on supporting health and well-being of school administrators, but psychological safety remains overlooked and is not part of the conversation. This calls for the need that psychological safety should be integrated into these health and well-being conversations. School administrators’ psychological safety has a ripple effect: A climate of psychological safety that invites open communication, risk-taking, and creative decision-making is not only vital for school administrators but also for teachers and students throughout the school community.