Individual differences in reactions to work location decisions

Purpose – The purpose of this study is to apply reciprocity theory to understand how hypothetical work location decision outcomes and individual differences affect employees’ trust in their employer and willingness to engage in organizational citizenship behaviors (OCBs). Design/methodology/approach – Three vignettes were used to manipulate work location decision outcomes and hypotheses were tested using Hayes’ (2008) PROCESS in a sample of 378 adults who worked in the USA during the first six months of the COVID-19 pandemic. Findings – Participants reported greater OCB intentions through higher trust in the employer when given their hypothetical choice of work location compared to being assigned one, and when assigned to their preferred compared to nonpreferred location. External work locus of control (EWLC) moderated the effects of work location on trust in the employer. The relationship between trust and OCB intentions was weakened when employees perceived greater difficulty in leaving their jobs. Originality/value – This study examined the roles of felt reciprocity, individual differences, choice and hypothetically receiving one’s preferred work location, on trust in the employer and willingness to engage in OCBs during the COVID-19 pandemic.

workplace: many employees now expect increased flexibility and have become more vocal about work location preferences (Barrero et al., 2021;Harter, 2020).
Relevant research has been conducted on flexible work arrangements including telework (Hill et al., 2008). Because this research has focused on voluntary remote work where employees' preferences are inherently reflected in their work location and is thus limited in explaining the unique situation during COVID-19, we propose a mechanism through which individual differences, work location choice and working at one's preferred or nonpreferred location affect employee trust in the employer and ultimately organizational citizenship behavior (OCB) intentions, during a time in which the employer may determine work location or allow employees to choose where to work. Applying the norm of reciprocity (Gouldner, 1960), we suggest that when employees perceive that their employer provides favorable treatment by letting them choose to work where they want, or by assigning them to work in their preferred location, trust in the employer increases which then increases employees' intentions to engage in OCBs (Aryee et al., 2002).
We also propose that because employees differ in their sense of control over their employment arrangement, they might react differently to being given a choice or being assigned a work location. For example, employees who believe work outcomes are controlled by external forces such as the employer (greater external work locus of control; Spector, 1988) might value work autonomy less and thus feel less obligated to reciprocate being given a choice. Similarly, employees with fewer alternative job opportunities, and thus feeling less control over their employment situation, may be less influenced by intrinsic motivation (i.e. trust in the employer) to engage in OCBs (Park, 2016). Difficulty of leaving which is similar to continuance commitment (Meyer et al., 1993) but more focused on one's perceived inability to leave the employer due to limited employment alternatives. Difficulty of leaving is relevant to employees' perceived control over their work environment, which is one of the key components of research on flexible work arrangements (Gerdenitsch et al., 2015). Thus, we investigate external work locus of control (EWLC) and difficulty of leaving (a component of continuance commitment; see Meyer et al., 1993) as moderators of the hypothesized mechanism.
This study makes several contributions to the literature. We add to the flexible work arrangements literature (Gerdenitsch et al., 2015;Hill et al., 2008) by extending the dichotomous approach to flexible work location decision processes (choice vs. no choice) and investigating new possible outcomes of this decision (working in one's preferred or nonpreferred location), reflecting the pandemic situation. We also respond to the call to consider individual differences in the flexible work arrangements research (Shockley & Allen, 2010) by assessing the roles of EWLC and difficulty of leaving the employer in these relationships. Figure 1 illustrates our conceptual model. The effects of adapting work arrangements to employee needs have received considerable scholarly attention (Hill et al., 2008). Flexible work practices satisfy the basic human need of autonomy (Deci & Ryan, 1995) by providing employees with control over their work location and schedule. Although some studies have warned of potential disadvantages including greater work-family conflict and work intensification (Gerdenitsch et al., 2015), providing a choice of work location is often considered an effective strategy to elicit positive work attitudes such as job satisfaction and organizational commitment (Kossek et al., 2006).
Letting employees choose where they work can give them more control over their work environment and allow them to better use their resources to effectively cope with stress and anxiety (Gajendran & Harrison, 2007), which might be particularly important during a pandemic. When offering this choice is not feasible, working in one's preferred location should at least help employees feel their needs and preferences (e.g. separation of work and life or flexibility to attend to family needs) are met. Indeed, research on needs-supplies fit (Kristof, 1996) and person-environment fit (Edwards & Shipp, 2007) has demonstrated the importance of a match between one's work environment and one's needs and its positive impact on employee attitudes toward their employer.
The reciprocity norm (Gouldner, 1960) proposes that giving employees resources they value not only satisfies their needs but also engenders a desire to reciprocate the gesture with higher trust and greater emotional engagement (Eisenberger et al., 2001). Accordingly, providing a choice of work location guarantees that the employee will work in their preferred space and initiates a social exchange relationship grounded in an environment of support, further reinforcing the trustworthiness of the employer (Aryee et al., 2002). An employee not having choice may still feel fulfilled and thus trust their employer more when assigned to their preferred work location because they are receiving an outcome they value. However, a mismatch between an employee's preferred and assigned work location should reduce trust in the employer because the employee is not receiving the outcome they value.
Furthermore, research has consistently reported a positive relationship between trust and OCBs (Moorman et al., 2018). Based on the reciprocity norm, we propose that being able to work in one's preferred location by choice or by employer assignment will be related to greater OCB intentions though higher trust in the employer. If an employer acts in ways beneficial to an employee and the exchange relationship is characterized by greater trust, the employee is likely to reciprocate with OCBs that are beneficial to the employer (Konovsky & Pugh, 1994). We suggest that providing work location choice is a voluntary supportive gesture on the part of the organization that in turn engenders trust and greater intrinsic motivation to engage in OCBs. When a choice is not given, an employee hypothetically assigned their preferred location will be more likely to intend to display OCBs through trust in their employer than an employee assigned to their nonpreferred location because their work location decision still provides an outcome they value: H1. Trust in the employer will mediate the relationship between hypothetical work location condition and OCB intentions.
External work locus of control EWLC relates to employees' expectations of the appropriateness of employer control over work outcomes, including work location assignments. People with a higher EWLC tend to believe that their work outcomes are determined not by their own choices and efforts but by powerful outside forces, such as luck or their employer (Lefcourt, 2014). Conversely, a lower

Reactions to work location decisions
EWLC is related to a higher sense of personal agency and sense that one has control over one's work outcomes (Spector, 1988). We suggest that employees will respond differently to hypothetical work location decisionswhether they are given a choicedepending on their EWLC. Because employees with a higher EWLC expect their employer to make workplace decisions and value autonomy less than those with a lower EWLC, they may therefore trust their employer less if the employer delegates this decision to them. In contrast, employees with a lower EWLC will have greater trust in their employer when they can choose where to work as this allows them to maintain a sense of control over their work situation. When a work location is assigned, employees with a higher EWLC may have lower trust in the employer if assigned their nonpreferred location because they are more likely to hold the employer responsible for this negative work outcome (Aubé et al., 2007): H2. EWLC will moderate the relationship between hypothetical work location condition and trust in the employer, such that employees with higher EWLC will show greater trust in the employer (a) when assigned to a hypothetical work location rather than when given a choice and (b) when assigned to their hypothetical preferred rather than their nonpreferred work location.
Difficulty of leaving the employer Lastly, we propose that the perceived difficulty of leaving will moderate the relationship between trust in the employer and OCB intentions such that the positive relationship between trust and OCB intentions will be weaker among individuals perceiving greater difficulty leaving their employer, and stronger among individuals perceiving lower difficulty of leaving their employer. As noted by Meyer et al. (1993), continuance commitment reflects both a difficulty in leaving due to limited employment alternatives and an investment in employer-specific skills that would be lost if one were to leave their current organization. Rather than investigating both components of continuance commitment, the focus of this study is on the effects of the perceived difficulty of leaving rather than on the value of investments made in the current employer because the lack of job alternatives should be the main driver of difficulty of leaving during the pandemic (Ansell & Mullins, 2021). Despite historic employment declines, some employees are still more mobile than others due to professional qualifications, work experiences or local unemployment rates (Ng et al., 2007). If employees think they can easily find new jobs, their trust in their employer should be more important in deciding to engage in OCBs. Conversely, if employees feel "stuck" with their employer, they may be willing to exert extra effort to avoid losing their jobs regardless of their perception of the employer. Relatedly, previous research has found that OCBs are higher when employees have fewer job alternatives (Thau et al., 2004): H3. Perceived difficulty of leaving will moderate the indirect effects of hypothetical work location condition on OCB intentions via trust in the employer, such that the positive relationship between trust in the employer and OCB intentions will be stronger when difficulty of leaving is lower.

Method
Sample and procedure We collected our sample through Qualtrics, a survey service platform that maintains panels of working adults with a broad range of demographic characteristics. We worked with the OMJ company to set recruitment criteria and included the screening questions in the first page of the survey (e.g. age and employment status). We restricted our survey to employees who continued working during the first six months of the COVID-19 outbreak to ensure that participants could easily relate their assigned vignette to their current job. Approximately 40% of invited participants completed the survey. After giving their assent to participate, participants answered questions about their preferred work location and individual differences (i.e. EWLC and difficulty of leaving), and then were randomly assigned one of the three vignettes (Appendix) describing a hypothetical work location announcement from their current employer. They then responded to questions assessing their trust in the employer and OCB intentions while thinking of the vignette they just read. Our final sample size was 378 (57% female; average age 48.5) and comprised 74% Whites, 11% Blacks, 9% Asians and 5% other. The mean organizational tenure was 12.11 years and participants had a variety of entry-level (12%), intermediate level (37%), first-level management (17%), middle-level management (21%) and senior level management (12%) jobs.

Manipulations
Three vignettes asked participants to imagine that they had received an email from their current employer announcing their work location from that point forward during COVID-19. Participants were randomly assigned to the "choice" (N = 123), "home" (N = 114), or "company" (N = 141) conditions.

Measures
All measures were rated using a seven-point Likert scale (1 = strongly disagree to 7 = strongly agree) unless indicated otherwise. Choice and preference. Because we aimed to investigate the effects of whether participants were given a hypothetical work location choice and whether the assigned hypothetical work location matched their preference on their work attitudes, we created two antecedents: choice and preference. Choice contrasts the choice group against the no choice group and preference contrasts the preferred group against the nonpreferred group. To create the antecedents, we used a coding strategy adopted by Garcia et al. (2010). As described above, the participants were randomly shown one of the three vignettes (choice, working at home and working at the employer's location) and asked to respond to the survey questions after reading the vignette. Then, just for the no choice group (those who were told to work at home, or at the employer's location), we coded participants into preferred and nonpreferred groups based on whether they were assigned the location they preferred. Individual work location preference was measured using the question, "If you had to choose one or the other, would you prefer to work from home or at your employer's location?" When participants who preferred working from home rather than at the company's location were assigned to the home condition, they were coded as preferred. Participants who preferred working at a company's location and assigned to the company condition were also coded as preferred. Unmatched cases were coded as nonpreferred. At this point, 123 individuals were in the "choice" group, 144 were in the "preferred" group and 103 were in the "nonpreferred" group.
Next, we contrast coded these responses to create our antecedents because dummy coding only enables comparisons between the choice condition and the preferred or nonpreferred condition (Cohen et al., 2003). Specifically, for choice, we coded those in the choice group as À2/3 and the preferred group as 1/3 and the nonpreferred group as 1/3. Preference was created by coding those in the choice group as 0, the preferred group Reactions to work location decisions as À1/2, and the nonpreferred group as 1/2 [for detailed review on creating multicategorical antecedents with contrast coding, see Hayes (2018, p. 407)].
External work locus of control. We measured EWLC using eight items from Spector (1988). A sample item is, "Promotions are usually a matter of good fortune". (a = 0.91).
Difficulty of leaving. Difficulty of leaving was measured as the participants' perceived difficulty of leaving their current employers and assessed with the five items of Meyer et al.'s (1993) continuance commitment scale relating to the difficulty of leaving due to the lack of job alternatives. Sample items include "I feel that I have too few options to consider leaving this organization". and "It would be very hard for me to leave my organization right now, even if I wanted to". (a = 0.80).
Trust in the employer. We used four items from Robinson and Rousseau (1994). A sample item is "My employer is being open and upfront with me". (a = 0.88).
Organizational citizenship behaviors intentions. OCB intentions were measured with seven items from Williams and Anderson (1991) and one item, "think of ways to do my job better", from Lehman and Simpson (1992) on a seven-point Likert scale (1 = extremely unlikely to 7 = extremely likely). (a = 0.85).
Manipulation check.
Participants were asked to indicate whether they agreed with the statements, "My company has given me a choice to work from home or at the company's location", "My company has required me to work from home (or at the company's location)". Participants whose responses did not correspond with their assigned conditions were stopped from continuing the survey and not included in any analyses. Table 1 presents the means, standard deviations, reliabilities and intercorrelations among the study variables. Confirmatory factor analyses showed that a four-factor model had an acceptable fit, x 2 = 526.93, df = 266, p < 0.01, CFI = 0.94, SRMR = 0.05 and RMSEA = 0.05, yielding a better fit than all other possible models [1].

Results
We used hierarchical regression and PROCESS (Hayes, 2018) to test hypotheses and grand-mean centered EWLC, trust in the employer and difficulty of leaving when testing Notes: N = 378. Cronbach's alphas are reported in parentheses on the diagonal. Choice and preference are contrast coded variables. **p< 0.01, *p< 0.05, two-tailed OMJ moderating effects. Age, gender and race were controlled as they are known to affect our study variables [2]. As shown in Table 2, trust in the employer was significantly related to work location choice (b = À0.43, p < 0.01) and preference (b = À0.45, p < 0.01), indicating that trust in the employer was significantly higher when participants were given a hypothetical choice than when assigned a work location, and when participants were hypothetically assigned to their preferred location than their nonpreferred location. Further, trust in the employer was significantly related to OCB intentions, b = 0.54, p < 0.001 (Model 3 in Table 2). To test for mediation of this relationship by trust in the employer, we further conducted bootstrapping to obtain confidence intervals (CIs) using PROCESS (Preacher & Hayes, 2008). The partially standardized relative indirect effects of choice and preference on OCB intentions via trust in the employer were À0.23 and À0.24, respectively and both 95% CIs did not include zero, [À0.369, À0.102], [À0.407, À0.088]. Thus, H1 was supported.
We also found support for H2a and H2b. The interaction between choice and EWLC was significant, b = 0.30, p = 0.005. Although the interaction between preference and EWLC was not significant, b = À0.09, p = 0.42, the highest order unconditional interaction was significant, DR 2 = 0.023, F(2, 361) = 4.48, p = 0.012, indicating that work location condition interacted with EWLC in predicting trust in the employer (Hayes, 2018) and that the moderating effect of EWLC explained 2% of the variances in OCB intentions. We plotted the interaction ( Figure 2) and examined whether the gap between the choice line and the midpoint line between preferred and nonpreferred groups, which reflects the contrast between choice and no choice conditions, varies across levels of EWLC (Hayes, 2018). Employees with lower EWLC showed much less trust in the employer when they did not have a choice in work location. For employees with higher EWLC, there was no significant difference in trust in the employer between when they were given a hypothetical choice and when they were assigned. For the interaction between preference (preferred vs nonpreferred) and Reactions to work location decisions EWLC, those in the preferred location condition had greater trust in the employer when their EWLC was higher while those in the nonpreferred location condition had lower trust in the employer when their EWLC was higher. Because employees with higher EWLC believe that it is their employer's responsibility to make good decisions for them, when assigned to nonpreferred work locations they might feel that their employer has reduced the value of the employment relationship to them and compromised their trust in the employer through negative reciprocity even when the assignment is hypothetical. Lastly, the proposed interaction between trust in the employer and difficulty of leaving accounted for 1% of the variance in OCB intentions and was statistically significant, b = À0.06, p = 0.02. The index of moderated mediation was 0.02 and its 95% CI did not include zero, [0.004, 0.054]. The interaction effect is shown in Figure 3. The relationship between trust in the employer and OCB intentions was stronger among participants perceiving a lower difficulty of leaving than among those perceiving a higher difficulty of leaving, supporting H3.

Discussion
The COVID-19 pandemic posed a unique set of challenges such that letting employees choose where they work may not be universally valued by employees or generate expected outcomes. This study examines these nuanced effects and finds that although having choice of work location matters, even if it is hypothetical, it does not always increase employees' trust in the employer and ultimately OCB intentions, as EWLC and the perceived difficulty of leaving moderate these relationships.

Theoretical implications
Providing choice in work location decisions generally enhances positive employee attitudes and behaviors (Gerdenitsch et al., 2015), although research findings on its impact on OCBs have been mixed (ter Hoeven & van Zoonen, 2020). In this regard, our findings advance  (Bakker & Demerouti, 2007;Hill et al., 2008) by revealing that trust in the employer is an important attitudinal mediator in linking work location decisions to OCB intentions. As major HR policies are often determined in a top-down manner, granting a choice or accommodating employee preferences in the decision-making process should have a direct bearing on employees' perception of their employer. In addition, trust is essential in promoting extra-role behavior (Asgari et al., 2008) and best achieved through employee participation and empowerment (Morgan & Zeffane, 2003), which are key components of giving employees a choice or assigning work locations based on employee preferences. However, previous research has tended to focus on employee-oriented outcomes such as job satisfaction (De Menezes & Kelliher, 2011) or work-life balance (Byron, 2005).
Our findings demonstrate that individual differences also play an important role in the process by which hypothetical work location decisions affect employee trust in the employer and OCB intentions. For employees with a higher EWLC in our study, whether they had control over work location decisions or not did not significantly affect their trust in the employer. When hypothetically assigned their nonpreferred location, employees with a higher EWLC reported significantly less trust than those with a lower EWLC because they hold stronger views that the employer should protect their interests and are hence more likely to hold it against the employer if their work environment does not meet their needs (Aubé et al., 2007).
It is also noteworthy that employees' translation of trust in their employer into OCB intentions differs depending on their perceived difficulty of leaving. Although trust has been found to be a critical precursor to OCBs (Moorman et al., 2018), it may have less of a positive effect for employees who feel stuck in their jobs, which might be an increasingly common scenario during the COVID-19 pandemic. Together, our findings add to recent efforts to incorporate individual differences into flexible work arrangements research (Shockley & Allen, 2010).

Practical implications
These findings have important implications for management practice. Although the COVID-19 pandemic has made it increasingly important for employers to accommodate employees' Reactions to work location decisions desires for flexibility and work location choice, this may not always be a viable option. Our findings highlight the importance of managerial awareness of employees' individual differences in maximizing the benefits of work location decisions. Not all employees value autonomy and the opportunity to have control over their work environment. In addition, the finding that participants with higher trust in the employer showed greater OCB intentions when they perceived lower difficulty of leaving while those with lower trust in the employer showed greater OCB intentions when they perceived greater difficulty of leaving implies that lack of control over employment (an extrinsic motivation factor) becomes more important to OCB intentions when employees' trust in the employer (an intrinsic motivation factor) is low. Enhanced OCB intentions are an important benefit of enhanced employee trust in the employer, particularly among employees who would have an easier time leaving the employer and who are likely more valuable employees as a result.

Study limitations and future research
Along with its academic and managerial implications, the study has a few limitations. First, the constructs were assessed with self-report surveys. Although this may raise concerns about common method variance (Podsakoff et al., 2012), a single-factor test (Harman, 1976) and a marker variable analysis (Lindell & Whitney, 2001) revealed that common method variance was unlikely to have had a significant impact on the results. Our results should also be interpreted with caution due to the use of cross-sectional data and the limitation of our research design created by the use of hypothetical vignettes. Our supplemental analyses showed that the participants' responses were not influenced by their actual availability of working from home and employment type (full vs part time). However, it should be noted that the generalizability of our results might be limited due to the nature of our research design. Future research may investigate additional mediators of these relationships. It is possible that different work location decisions affect employee socialization (Bauer et al., 2007) or knowledge sharing (Ipe, 2003) as employees are exposed to different levels of interactions when they are allowed to choose their work locations compared to when they are required to work at a designated place. Our findings shed light on workplace management during the COVID-19 crisis and answer calls for a more nuanced approach to the use of flexibility in work locations for better workplace outcomes. Notes 1. The results of confirmatory factor analyses on three-factor, two-factor and one-factor models are available from the corresponding author upon request.
2. Because participants' actual availability of working from home and their employment type (partor full-time) could have affected their reactions to the assigned vignettes, we conducted ANOVA and found no significant differences. We further tested our hypotheses controlling for the availability of working from home and employment type and found that the results remained unchanged.

Employee choice condition
Thank you again for all that you've done during the COVID-19 upheaval. Your continued efforts and resilience have helped us to be well positioned to move forward successfully. I know that it has been a very challenging time, and I have been very proud of how our company has handled this unprecedented situation. I want to update you on the company's decision regarding work locations.
We are now giving all employees the option of working from home or from the office. We trust that you will make the decision that is best for you and your family, and your choice will be honored without penalty. Please let your supervisor know your choice by the end of the day next Friday. If you have any questions, please feel free to contact your supervisor.
Working from home condition Thank you again for all that you've done during the COVID-19 upheaval. Your continued efforts and resilience have helped us to be well positioned to move forward successfully. I know that it has been a very challenging time, and I have been very proud of how our company has handled this unprecedented situation. I want to update you on the company's decision regarding work locations.
We are now requiring all employees to work from home beginning next Friday. We think this is in the best interest of our customers, employees and business. If you have any questions, please feel free to contact your supervisor.
Working at the company's worksite condition Thank you again for all that you've done during the COVID-19 upheaval. Your continued efforts and resilience have helped us to be well positioned to move forward successfully. I know that it has been a very challenging time, and I have been very proud of how our company has handled this unprecedented situation. I want to update you on the company's decision regarding work locations.
We are now requiring all employees to return to work at the company's location beginning next Friday. We will provide protective equipment and sanitizing stations to protect our employees and customers. If you have any questions, please feel free to contact your supervisor.