A qualitative exploration of trust in the contemporary workplace

ABSTRACT Objective The Australian work environment is changing rapidly, as employees and leaders are becoming used to a blend between a virtual and face-to-face workplace. In this context, this research aimed to explore how leaders build and sustain trust in the contemporary workplace. Method Critical incident technique was used in interviews with employees and a constructivist grounded theory approach via reflexive and coding thematic analysis was used to interpret the results. Thirty-eight participants were interviewed based on purposive sampling from virtual and face-to-face work environments and were asked about their experiences with trust in direct managers and senior leaders. Results Employing an inductive approach, this research found an explicit behavioural component and an interpersonal relationship component, thus reinforcing organisational trust multidimensionality. The role of communication, exposure and relationships were highlighted as critical in virtual work environments, regardless of leader type. Conclusion Using a qualitative research design enabled greater clarity of what constitutes organisational trust in the contemporary workplace. Implications for organisational trust theory are also discussed. KEY POINTS What is already known about this topic: Trust in the workplace has been difficult to define. Trust in the workplace is essential for organisational performance. Trust between employees and leaders is of particular importance and worth examination. What this topic adds: The work environment has changed forever, and this influences how trust is developed and maintained between employees and leaders. Trust in the contemporary work environment of a blended virtual and face-to-face context comprises a behavioural and relational component. Exposure, communication and strong relationships between employees and leaders are essential for trust in a virtual work context, regardless of leader type.


Introduction
The nature of work, leadership and teams has changed due to globalisation, technological improvement, increasing cost for space, and now employee safety and infection control due to the COVID-19 pandemic in Australia and beyond (Atkinson & Hall, 2011;Boddy, 2020;Galea et al., 2014). The pandemic thrust organisations into the virtual workplace like no event has done before (Graham, 2020;Rudolph et al., 2021;Shine, 2020), and there is strong evidence that Australian workplaces will remain a blended virtual and face-to-face work environment (Productivity Commission, 1 2021).

The conceptualisation of trust in the workplace
The academic literature confirms how trust is fundamental to quality leadership. Trust is the bedrock for many employee outcomes, such as job performance and satisfaction (Zhu et al., 2013), organisational commitment , organisational citizenship behaviour (Mushonga et al., 2014) and psychological safety (Frazier et al., 2017). Yet, historically, trust has been given multiple definitions, and the literature considers it a trait, a state and a cognitive process Schoorman et al., 2007). More recent research has used Rousseau et al. (1998) and Schoorman et al. (2007) definitions that describe trust as a state of vulnerability to another person (Agote et al., 2016;Chen et al., 2012). Others have defined trust as interpersonal risk, expectations of another's benevolence, and an attribution of positive intentions of the words or actions of others (McEvily & Tortoriello, 2011). McEvily and Tortoriello (2011) found 129 unique measures of organisational trust, based on any one of these definitions, in their review of organisational trust measurement, and noted the little replication between these measures.
Although research on organisational trust is in a mature state, a meta-analysis by Dirks and Ferrin (2002) highlighted that trust may not be unidimensional, but rather multidimensional. Two reviews of trust measurement (Dietz & Den Hartog, 2006;McEvily & Tortoriello, 2011) and Schoorman et al. (2007) "Integrative Model of Organisational Trust" reinforce trust as multidimensional. The most common nomenclature for the dual dimensions of trust is affective and cognitive trust (Carter & Mossholder, 2015;Miao et al., 2014;Newman et al., 2014). Affective trust is dependent on the quality of relationships, whilst cognitive trust forms from the leader's displayed behavioural integrity (Dirks & Ferrin, 2002;McAllister, 1995;Schaubroeck et al., 2011). Regarding the measurement of multidimensional organisational trust, McAllister (1995) scale is the principally used measure (McEvily & Tortoriello, 2011). At the time this paper was written, Google Scholar indicated that there are nearly 10,000 citations of McAllister (1995) study.
More recently, a meta-analysis investigating the relationship between trust dimensions and organisational citizenship behaviour and organisational commitment examined these relationships directly and found significant differences between trust dimensions and organisational outcomes, further strengthening the theory that trust is multidimensional (Fischer et al., 2020). The literature has called for a review of the terminology used to describe multidimensional organisational trust, as well as a review of the scales used to assess trust between employees and leaders (Dietz & Den Hartog, 2006;Dirks & Ferrin, 2002;Fischer et al., 2020;McEvily & Tortoriello, 2011). This paper attempts to address a part of that need by creating clarity about trust in the contemporary work environment.

Contemporary workplace nuances that could affect conceptualisation of trust
An additional consideration about conceptualising trust in the workplace relates to differences between the virtual, hybrid and face-to-face workplaces. That literature is growing and there has been some research on the importance of trust building in teams led virtually (Caulat, 2012;Lauring & Jonasson, 2018). Antonakis and Atwater (2002) described this emerging organisational phenomenon twenty years ago and the challenges it presents for trust development. Ferazzi (2012) noted that establishing trust in face-to-face workplaces is challenging enough, and even more so in a virtual workplace where individuals are not likely to meet in person. There is some evidence that the trust-building process between employees and virtual leaders is similar to the process in a face-to-face leader-employee relationship, but there are also potential differences that must be explored (Hirschy, 2011;Malhotra et al., 2007;Merriman et al., 2007;Oertig & Buergi, 2006).
Finally, leader type appears to play an important role in how trust influences outcomes in the workplace. Dirks and Ferrin (2002) identified the important distinctions that individuals make between different leadership referents (e.g., direct manager versus senior leader of organisations) and the implications of these distinctions. The distinction employees make in how trust forms and how different leader referent groups sustain trust may provide understanding of how organisational leaders can practically build trust with their employees to improve outcomes. Dirks and Ferrin (2002) found evidence that employees trust direct managers and senior leadership in distinctive ways and that trust in these two referent groups, when explored as separate entities, influences different organisational outcomes. These findings suggest that trust has a differing relationship with antecedent and outcome variables and that the relationship between employees and direct managers versus senior leadership differs in relation to trust.

The role of qualitative research in theory development
Qualitative methods have a key role in theory development (DeVellis, 2016). Even at the time of development of some of the foremost theories and measures of trust, the use of focus groups was a common practice before developing theory and structured questionnaires (O'Brien, 1993;Steckler et al., 1992). Contemporary scale development practices to assess theoretical constructs consider lack of qualitative research a limitation (Morgado et al., 2018). Although there is a case for qualitative research in theory development, it is not immune to risk, due to researcher subjectivity. When selecting a qualitative research method, it is important to choose a method with the greatest chance to reduce subjective bias (Morgado et al., 2018). Critical incident technique (CIT) is a useful qualitative data collection tool for theory and scale development (Dugar, 2018), whereby critical information about behaviour in specific situations is identified and analysed (Flanagan, 1954). This method is a widely used form of task analysis in organisational psychology, but it is also an investigative tool used to explore psychological constructs or experiences (Butterfield et al., 2005). The CIT method reduces subjective biases from both the researcher and the participant (Gremler, 2004). The CIT approach asks participants to recall and describe their positive and negative experiences, as critical incidents, in relation to a topic of interest rather than focussing on generalisations or opinions (Byrne, 2001;Walker & Campbell, 2013). The recalled situation is considered a critical incident if the participant's description includes the setting in which the incident occurred, the behaviour of those involved, including attitudes, emotions or actions, and the outcome of the behaviour (Walker & Campbell, 2013).
The choice of qualitative analysis method is also an important consideration. Thematic analysis (TA) is a useful qualitative data analysis technique for theory development, as it allows for flexible, iterative review of the data to arrive at the results (Choudhary et al., 2018). TA encompasses a variety of approaches and each approach has a nuanced rationale for application (Braun & Clarke, 2021a). TA literature describes how reflexive TA can be used for theory development, as it can be used inductively and to interpret complex subjects (Braun & Clarke, 2019). Whereas coding reliability TA is a more methodical and deductive approach and attempts to reduce bias and increase quality assurance in the data analysis (Braun & Clarke, 2020).

Research rationale
The purpose of this investigation was to explore organisational trust in the contemporary work environment using qualitative research methods to determine if the virtual or face-to-face work context has had an influence. The researchers' broader intention was to examine validity of the range of existing theoretical perspectives about trust in the contemporary workplace which have informed measures used in ongoing research. Time has passed since the literature examined the conceptualisation of organisational trust (Dirks & Ferrin, 2002;Schoorman et al., 2007). Current trust literature may not be capturing the full scope of organisational trust to date and there have been no qualitative, empirical attempts to bridge these theoretical development gaps (Dietz & Den Hartog, 2006;McEvily & Tortoriello, 2011). It is also possible that current organisational trust measures used in research are not necessarily reflective of the present-day nature of employee-leader contexts, such as virtual leadership and agile work practices. The COVID-19 pandemic caused workplaces to change how work was conducted, based on new restrictions aiming to increase infection prevention and control (McLaren & Wang, 2020). Teams that would never imagine working virtually switched into a remote working environment, and their previous physical boundary condition ceased to exist (Arruda, 2021;Graham, 2020;Power, 2020;Rudolph et al., 2021;Shine, 2020).

Research aims
This study was undertaken to explore employee perceptions of trust in their leaders and included specific identification of direct managers and senior leaders as well as participant sampling according to virtual or face-to-face work environment. CIT was used for data collection and thematic analysis was used to interpret the responses. Due to the timing of the study, one cohort of participants was sampled before forced virtual working prior to the COVID-19 pandemic and one cohort was interviewed during the pandemic. This allowed determination of differences in employee trust in leaders directly impacted by forced separation. Therefore, the aims of this study were to explore: (1) What constitutes employee trust in leaders in the contemporary workplace, (2) Whether employee trust in virtual leaders differs from face-to-face leaders, and (3) Whether employee trust in direct managers differs from senior leaders.

Theoretical approach
A qualitative research design using a constructivist grounded theory approach was used to explore the research aims. Constructivist grounded theory is a research method focused on the generation of theory that is identified from data that has been systematically collected, compared and analysed (Charmaz, 2006). This process is also known as the inductive method of analysis (Azungah, 2018;Noble & Mitchell, 2016). Researchers use this approach to gain understanding from participants' responses and provide explanations to the research questions, rather than attempting to prove or disprove a theory (Charmaz, 2014;Mills et al., 2006). Critical incident technique (CIT) as the data collection method elicited positive and negative examples of employee trust in leaders. A combination of reflexive and coding reliability thematic analyses (TA), two methodological tools of a constructivist grounded theory approach (Braun & Clarke, 2021b), was used to analyse data. Repeated statements and descriptions became apparent and were coded. Codes were then grouped into concepts, and finally into themes and subthemes. Authors used the COREQ checklist (Tong et al., 2007) to ensure robust reporting of the study design and analysis of findings. This study employed reflexive TA and coding reliability TA for inductive purposes about the construct of organisational trust (Braun & Clarke, 2019). Initially, a reflexive TA approach was used. Reflexive TA is flexible and organic, and analysis output evolves throughout the coding process, like initial coding in grounded theory (Braun & Clarke, 2020;Charmaz, 2014). Repeated statements and descriptions became apparent and were coded. Codes were then grouped into concepts, and finally into themes. After applying reflexive TA, coding reliability TA was employed to determine saliency in the theme. As explained, the trust literature is not unified and so taking a methodical, yet open approach to the data was essential so as not to miss crucial information or allow researcher bias to place undue focus on any one part of the data.

Participants
Participants were 38 employees (15 men, 23 women) aged 30-55, who worked full-time, held a Bachelor's degree or higher, and were employed in occupations in industries such as law, finance and banking, design, recruitment, marketing, advertising, government, other public service, and consulting. Participants worked from a range of locations in Australia (N = 27), as well as locations in the United States (N = 6) and United Kingdom (N = 5). Participants were also grouped based on whether they worked in teams virtually (n = 29) or face-to-face (n = 9). 2 A portion of the participants (55%) were interviewed during a time of forced virtual working due to government restrictions to ensure infection prevention and control of COVID-19.

Materials and procedure
Participants were recruited via advertisements placed on social media sites, such as LinkedIn and Facebook. The snowball sampling method was employed through researchers', as well as participants', personal and professional networks, such as volunteer organisations, MBA programs, professional associations, workplaces, etc. Participation involved discussing personal experiences trusting a direct manager and a senior leader that was at least two levels above in the organisational hierarchy. Selection bias was decreased by ensuring that participants came from a variety of corporate and government occupations in Australia, United States and United Kingdom.
Interviews occurred in a mix of face-to-face and virtual meetings. Researchers conducted face-to-face interviews at the participants' place of work during normal work hours, whilst virtual interviews were scheduled at the participants' convenience over Skype or telephone. Before interviewing began, participants completed a consent form and a brief questionnaire of demographic items. Participants in virtual interviews were emailed the form and questionnaire in advance and were asked to return the completed form prior to interview. At the start of the interview, participants were reminded that interviews were being recorded. The interviews took approximately 30-45 minutes. Interviews were digitally recorded and then transcribed verbatim. Following transcription, identifying information was removed from transcripts to protect participants' anonymity, which further supported a reduction in the potential for researcher bias.
Using the CIT methodology (Flanagan, 1954) participants were asked to 1) describe a particular occasion when they trusted a direct manager and to 2) describe an occasion where their trust in a direct manager was broken. Participants were asked the same questions in relation to experiences of trust/broken trust with a senior leader, and specifically to a) describe what happened, b) how they felt, and c) what the outcome was. If the participant first provided a positive example in relation to trust with a direct manager or senior leader, participants were then asked to provide a negative example, and vice versa. The use of CIT inhibited participant social desirability bias and interviewer cognitive bias, because it specifically asked about personal experiences, both positive and negative, in relation to employee trust in leaders.

Analysis
A two-step analysis approach was conducted. First, reflexive TA was employed (Braun & Clarke, 2020) to review interview transcripts to derive the themes and subthemes related to employee trust in leaders. Then, line-by-line coding, also known as coding TA, was used to identify key words, phrases, and sentences, while paying specific attention to language use. From the codes, a set of three themes was developed, with subthemes added subsequently as the themes were deemed too broad. Coding was iterative and refinement of themes and subthemes evolved over the course of the analysis (Braun & Clarke, 2021b;Corbin & Strauss, 2008).
Next, double coding (Boyatzis, 1998) was used to establish inter-rater reliability. Establishing inter-rater reliability is a widely acknowledged method of ensuring the trustworthiness and quality of qualitative research (McAlister et al., 2017). Double coding involved the first author coding all interviews. Then a second researcher coded 30% of the data. To reduce potential bias, the second researcher had no prior involvement or review of the research project. The second coder was provided with a codebook to conduct their analysis. Inter-rater reliability was calculated using Cohen's Kappa (Armstrong et al., 1997), with an initial agreement level of κ = 0.86 achieved between the two raters. Kappa values above 0.80 suggest strong agreement (MacPhail et al., 2016). The raters discussed discrepancies and, where appropriate, themes and subthemes were refined until coders reached full agreement about the final version (Burla et al., 2008).
To examine the differences, if any, between virtual and face-to-face led employees and between employee trust in direct managers versus senior leaders, coders coded quotes accordingly. Coders coded responses about direct managers as "D" and senior leader responses as "S". Coders coded responses from a participant led virtually with "V" and face-to-face led participants as "F". Responses were also coded as "BP" and "DP" to indicate whether the interview occurred before or during the pandemic.
Several have attempted to answer the question of how many interviews are enough concerning theoretical saturation, with no unified solution (Aldiabat & Le Navenec, 2018). Researcher views range from 12 to 30 participant interviews (Francis et al., 2010;Guest et al., 2006;Mason, 2010) for adequate saturation. Theoretical saturation was reached at 30, but interviews continued up to 38 interviews as the subject matter required more certainty to ensure all themes and subthemes were captured.

Results
The analysis ultimately yielded three themes, with two to six subthemes per theme, relating to employee trust: trustworthy behaviours, interpersonal connection and care, and trust enablers. Researchers initially reviewed the pre-pandemic and during pandemic data sets separately to determine differences. No substantial difference was found between data sets regarding themes and subthemes. The combined results are reported according to the three themes and relevant subthemes. Each theme section includes a table with illustrative quotes characterising subthemes to represent participants' voice.

Trustworthy behaviours
This first theme related to a set of behaviours that leaders demonstrate and are observable to the employee. These behaviours related to honesty and integrity, communication, support and equality, and e2095226-6 respect and are described below. Authors did not note any differences in participants' responses regarding virtual versus face-to-face teams or leader referent group. See Table 1 for illustrative quotes of subthemes.

Honesty and integrity
Integrity is the quality or state of being honest. Participants used the words "honesty" and 'integrity' interchangeably to describe leader behaviour that appeared to enhance employee trust. Participants discussed the importance of being honest about what was happening or going to happen in their organisations during the time of uncertainty caused by the COVID-19 pandemic.

Communication behaviours
Communication behaviours comprised many elements. First, the mere sharing of information builds the foundations for trust according to participants. The second element was the ability to listen to, empathise with and/or demonstrate acceptance of the employee. Third, participants discussed how transparency positively impacts trust. When participants felt that information was being retained or hidden, it seemed to create insecurity. In contrast, transparency appeared to signal a level of comfort and so the employee was more likely to trust. This aspect of communication seemed to be about clarity of information. Finally, the fourth communication element was the ability for the leader to ensure confidentiality. Many participants described the importance of their leaders keeping employee information or concerns private. Further, participants said without face time with leaders and teams, employees and leaders had to put a greater emphasis on communication. The more communication from direct managers and senior leaders, the more participants felt they could trust despite the virtual environment.

Reliability
Participants described the behaviour of reliability as important to building trust. Reliability was used to articulate employee perception of the consistency of leader behaviour and that past behaviour could be counted on to support belief that the leader is trustworthy. Participants also discussed times when leaders could not be relied on to act according to their word, and how that broke trust.

Support
When leaders appeared to support their employees, trust was enhanced. Alternatively, when employees were not supported, trust was damaged. Different forms of support were described by participants. For example, instances where leaders did or did not support learning from mistakes impacted participants' ability to trust. Participants also discussed occasions when leaders advocated for their employee's position, rather than their own or with some other agenda. Support through acts of mentorship was also noted. The act of supporting the employee in times of need was another frequently described supportive behaviour that influenced trust positively. In the context of the pandemic, this was described as understanding employee physical and psychological wellbeing about infection. It also included support for employees who experienced increase carer responsibilities due to childcare facilities, schools, and aged care facilities being locked down.

Equality and respect
The final explicit behaviour described by participants related to equality and respect. Participants perceived that leaders who demonstrate respect and treat employees as equals, rather than as subordinates, increased trust. Conversely, participants remarked how disrespectful behaviour and instances where the leader used power dynamics either damaged trust or had a detrimental effect on trust development.

Interpersonal connection and care
The second theme that was identified in the data was about factors related to interpersonal connection and care between the employee and the leader. This may also be thought of as relationship quality. See Table 2 for illustrative quotes relating to the subthemes.

Exposure
Participants described exposure to the leader as an important aspect of relationship quality, and participant responses consisted of descriptions of both times spent in the relationship and frequency of contact with the person. It appears that the longer the employee spends in the employee-leader relationship, the more likely they are to observe trustworthy behaviour. Similarly, participants described how contact with their leader and the level to which the leader is visible impacts building trusting relationships. Participants' responses to this subtheme varied regarding whether they worked in virtual or face-to-face work environments. Those in virtual teams discussed how exposure is less organic and that more effort is required for leaders to establish and maintain workplace relationships. In a forced virtual working environment, exposure was mentioned by nearly all participants as important to building and maintaining trust. Regarding leader referent group, participants described senior leaders as less visible and, due to organisational hierarchy, there were not many opportunities for exposure to employees. Participants considered those leaders who were visible as trustworthy.

Rapport and understanding
Good rapport and the ability to familiarise and build understanding was identified by participants as essential to developing trust. It appeared to be the first step in establishing relationships between employees and leaders. Leaders who invested time in developing relationships with their employees positively influenced trust. Regarding the pandemic, some participants started new roles at this time. Building rapport was emphasised as essential in a virtual, new work environment.

Sameness
When employees and leaders were able to identify areas of common ground or similarities, a sense of sameness appeared to develop and thus trust development became easier. The ability to identify sameness between employee and leader seemed to be an enabler to develop rapport and understanding. Employees also described shared experiences as an additional important way to create sameness if common ground or similarities were not easily apparent in the relationship.
Participants also described how the pandemic gave each other new things to talk about and find common ground. Video conferencing allowed employees and leaders glimpses into their homes, which became useful information to develop new connections.

Leader vulnerability
Participants discussed situations when leaders displayed vulnerability to their employees that signalled that the leader could be trusted. When leaders openly displayed their vulnerabilities in the employee-leader relationship, it appeared to deepen employee feelings of trust. The same was true for employees to the leader. When employees felt they could trust their leader, then they were more comfortable to show their vulnerabilities.

Employee psychological safety
Participants described the importance of feeling safe and secure as employees in the relationship with the leader. When taking a risk, there were negative consequences for trust when psychological safety was compromised. The pandemic created a risky environment for employees, and participants consistently described the importance of feeling safe and secure at this time more so than prior to the pandemic. Participants described an increase in needing to be open with leaders about challenges at work and home and needing to trust leaders that they will not be penalised or judged.

Emotional connection
Participants also discussed feelings of emotional connection about their perceptions that trust exists, using words such as "care". Participants described experiences with leaders that were almost familial and protective, within the context of the workplace. For the cohort interviewed during the pandemic, this was particularly evident within senior leader examples. Participants emphasised the importance of the care and concern leaders demonstrated for their health and wellbeing during an uncertain time.

Trust enablers
The third theme identified from participant responses was about the specific enablers that support observation of trustworthy behaviours and factors related to interpersonal connection and care. The two enablers discussed were about self-propensity to trust and technology, and illustrative quotes of these subthemes can be found in Table 3.

Trust propensity
An employee's propensity to trust seemed to greatly impact how trust is built and maintained between employee and leader. Trust propensity appeared as a subtheme, as an enabler or barrier, based on the language participants used when prefacing or explaining examples of instances when they could or could not trust direct managers and senior leaders. Further, trust propensity seemed to differ for participants and may be based on individual experiences that affect the individual's belief or perception that trust exists.

Technology
For participants working in the virtual working environment, technology was an essential enabler to observing trustworthy behaviours and factors related to e2095226-8 interpersonal connection and care. Technology supports communication and exposure, which then influences the ability to demonstrate honesty, reliability, support, and respect, as well as build deeper relationships.

Discussion
The first aim was to explore what constitutes employee trust in leaders qualitatively through interviewing employees about their experience trusting leaders. Participants' description of trust experiences indicated trust is multidimensional, and that there are trustworthy behaviours and an importance placed on interpersonal connection and care for trust development and maintenance. The finding regarding organisational trust multidimensionality is consistent with McAllister (1995) and Dirks and Ferrin (2002) and endorses their contribution to the trust literature. Further, two other recent organisational trust reviews discussed similar trustworthy behaviours to what this study found (Dietz & Den Hartog, 2006;McEvily & Tortoriello, 2011). This paper upholds Rousseau et al. (1998) position on vulnerability, in addition to the relationship of temporality to trust (Schoorman et al., 2007). Trust's reciprocal nature and the significance of the emotional connection was evident in the results (Schoorman et al., 2007), as well as the relationship to trust propensity (Ashleigh et al., 2012;Dietz & Den Hartog, 2006). Trust propensity is an individual characteristic that explains a person's natural set point from which they build trust in an interpersonal relationship (Alarcon et al., 2018). The process is likened to an individual's deliberate comparison of the potential costs and benefits from a given set of decision options, from which the individual selects the most optimal course of action (Alarcon et al., 2018). Costa and Anderson (2011) model of multidimensional team member trust comprises trust propensity, perceived trustworthiness, and trust-building behaviours, which is consistent with the findings from this study. Dietz and Den Hartog (2006) also described the trust process in a similar way to this study. The authors frame trust (Dietz & Den Hartog, 2006) as a series of interactions between individuals' observations of trustworthy behaviour, the quality and nature of the relationship, and individuals' trust propensity, trust belief, and willingness to be vulnerable. This study reaffirms the importance of trust propensity and reciprocity in trust development (Costa & Anderson, 2011;Seppälä et al., 2011) and appears to be consistent with multidimensional organisational trust between team members (Costa & Anderson, 2011).
The second and third aims were to determine whether the perception of trust in leaders differed between groups. This included specific identification of direct managers and senior leaders and participant sampling according to virtual or face-to-face work environment. There were no differences in how participants from virtual and face-to-face work environments described the behavioural enablers of trust in leaders. Regarding leader referent group, participants did not suggest any behavioural differences either. However, those with limited exposure to their leader (e.g., working virtually or with limited access to senior leaders) noted that trust formation was less organic and that more effort was required for leaders to visibly display behaviour and to establish and maintain workplace relationships.
There was an interesting nuance that was noticed in participant interviews that were conducted with employees working during the pandemic. The number of positive examples of trust with senior leaders was higher than in the participant interviews conducted prior to the pandemic. Responses suggest this was due to increased senior leader communication in the virtual work environment with specific reference to the pandemic's influence on operations. The pandemic cohort of participants' examples showed an increase in subthemes of Communication Behaviours and Exposure to direct managers and senior leaders during an uncertain time through virtual and video meetings and email. This is where the Trust Enabler subtheme of Technology showed prevalence, rather than Trust Propensity. Technology was an important tool for leaders to gain exposure, communicate and show their care for employees' wellbeing rapidly and often. Lastly, Emotional Connection was emphasised much more from the participant cohort interviewed during the pandemic, because of the many critical incidents that participants shared about their leaders involved showing care through checking on physical and mental health and wellbeing and interest in coping with new burdens. The burdens included working whilst home-schooling children, caring for pre-school aged children when day-cares were closed, and increased hygiene efforts to protect vulnerable family members.
This suggests reframing thinking about work environment and leader type to consider leader social distance as the influencing factor, rather than leader type or work environment. The literature acknowledges leader social distance as a factor in how trust develops (Antonakis & Atwater, 2002). These findings reinforce Antonakis and Atwater (2002) assertion that leader social distance helps to explain leader-influencing processes and, by extension, how trust with the leader develops.

The significance of qualitative research methods in theory development
Support for multidimensional organisational trust theory could not be upheld without the use of qualitative research methods. The CIT method afforded participants the conduit to describe trust within real contexts, grounded in experiences trusting and mistrusting leaders. The two thematic analyses enabled inductive response interpretation from the point of view of the participants. Qualitative research is also a useful tool to ascertain relevancy of scales to measure constructs, as environmental contexts change over time (DeVellis, 2016). These findings also extend multidimensional organisational trust theory by providing hidden detail within the behavioural-based and the relationship-based dimensions, as well as highlighting a process through which these dimensions create the psychological state of trust. This could only be achieved through employing the essential qualitative research step in theory development (DeVellis, 2016;O'Brien, 1993).

Limitations
This study has some limitations, which should be acknowledged. The themes and subthemes found in participants' interviews reflect the experiences of the individuals sampled. Those participants being from a Western, democratic society with a potentially higher than average level of education and socioeconomic status. This was due to the sampling strategy chosen, as it was not a randomised group of participants. Although the participants were from a range of professions, it is not assumed that these outcomes are generalisable to employee-leader relationships in all work environments in all countries, and therefore the theory must the thoroughly tested across populations.
Another limitation of this study is the potential for researcher subjectivity bias and participant social desirability bias, given the snowball sampling method that included researcher and participant personal and professional networks. This was partly resolved using CIT. The process of double coding to establish reliability was also employed as a check for researcher subjectivity bias, as the second coder was completely isolated from the research project until coding and themes were refined through discussion as part of the process.
Finally, at the time of drafting this paper, the rate of literature on the impact of the COVID-19 pandemic was increasing. There may have been new information available in the academic community that could have been considered or referenced but was not available to the authors at the time.

Future research
This study is a further step in organisational trust theory development, and this could not be achieved without using qualitative research methods. Organisational trust appears to be multidimensional and has a behavioural and a relational dimension. Employees develop and maintain trust through continual observation of trustworthy behaviours and benchmarking those behaviours within the context of a quality relationship. The trustworthy behaviours of honesty, communication, reliability, support, equality and respect are consistent with previous research (Dietz & Den Hartog, 2006;McEvily & Tortoriello, 2011). The literature also describes the importance of exposure and leader vulnerability Schoorman et al., 2007) to trust, but this study extends our understanding about the relational dimension and the factors that underpin interpersonal connection and care that drive trust. Further, these the findings regarding leader vulnerability and employee psychological safety support Rousseau (1998) theory of the psychological contract within the context of a trust relationship. This would benefit from further exploration. Transformational Leadership Theory (Zhu et al., 2013) also discusses the role of the leader behaviour and ability to build strong relationships with followers to achieve outcomes, which this study reinforced as key for building trust at work. For example, Zhu et al. (2013) found that employee affective trust in leaders translated transformational leadership into positive organisational outcomes. This study prompts further investigation into the advantage of Transformational Leadership in organisations in building organisational trust, particularly in this new world of hybrid face to face and remote working.
The intention of the paper was to explore the leader-employee trust relationship in the traditional and virtual workplace or organisation to fill a gap in the literature (Dirks & Ferrin, 2002;McAllister, 1995). The contingent workforce, or what is commonly referred to as the "gig economy" (Kaine & Josserand, 2019), was deemed out of scope for this study as these roles do not consistently fall into the 'organisational' context. However, employee-employer trust would be an excellent area of focus for future research.
As noted, many papers on organisational trust theory development and measurement do not include detail on the critical qualitative research step. Moreover, the working climate has shifted since many seminal scales were developed with the rise of virtual teams, agile workforce, and the blurred line between home and work (Ferazzi, 2012;Graham, 2020;Rudolph et al., 2021;Shine, 2020). Therefore, the validity of organisational trust measures based on previous organisational trust theory needs to be confirmed for the current working climate considering this paper's findings. The most prominent measure of relationship-based and behavioural-based trust is McAllister (1995) scale (Fischer et al., 2020;McEvily & Tortoriello, 2011), however, the paper bases its scale development on an expert review of items by "eleven organisational behaviour scholars" (page 35) and not a qualitative exploration of the construct. McAllister (1995) measure should be reviewed and adapted, or a new measure could be developed, to reflect employees' perceptions of trust in leaders accurately. This measure must then be tested with a variety of workplace and cultural contexts to confirm generalisability.

The Productivity Commission is the Australian
Government's independent research and advisory body on a range of economic, social and environmental issues affecting the welfare of Australians. 2. Face-to-face working is considered as an environment where teams interact face-to-face on a regular basis.

Disclosure statement
Sarah Fischer and Arlene Walker did not receive any financial or other support. Sarah led the design, data collection, analysis, and interpretation, and writing of the article, under the supervision of Arlene Walker as part of Sarah's PhD. No conflicts of interest exist for any of the authors of this study.

Data availability statement
Due to the nature of this research, participants of this study did not agree for their data to be shared publicly, so supporting data is not available.