Keywords

1 Introduction

A long-standing tradition in virtual team (VT) literature is the adoption of the proximity perspective to identify the unique features of VTs and their impact on individual dynamics and group processes. Researchers have conceptualized multiple dimensions of “virtuality” (e.g., degrees of spatial, temporal, and cultural distances) to reflect the proximal characteristics of a VT and examine how virtuality influences individual variance in work outcomes. However, the empirical results have been inconsistent, with studies showing positive and negative relationships between proximity and work outcomes. Some studies found that the objective distance between individuals negatively influences communication quality, task performance, and knowledge sharing [2, 9]. In contrast, some studies have shown the positive influence of objective distance on work outcomes, such as decision quality, creativity, and job satisfaction [22, 34].

To resolve this inconsistency, scholars have recently called for research to address subjective representations of distance between VT members, instead focusing on objective distance measures [13, 28, 35, 36]. Trope et al. [31] have argued that psychological distance is a factor that leads individuals to create higher level abstractions in the representation of target objects (“higher-level construals”). These higher-level construals disrupt and prevent information flow between virtual workers [36]. If psychological distance is an underlying mechanism that accounts for the representation of VT work and work outcomes, providing information technology (IT) functions to mitigate the objective distance may not be the most suitable approach. Instead, technological and managerial interventions should be focused on changing team members’ perceptions of the subjective, psychological distance. Lesser psychological distance should lead to lower-level construals and more effective VT work. Yet despite the increased attention from research and practice, the operational conceptualization of psychological distance remains equivocal; research findings regarding its roles on work outcomes are inconsistent and often appear conflicting. To fill these gaps in understanding, this study addresses the following questions:

  1. (1)

    How is the psychological distance between virtual workers formed?

  2. (2)

    How does the psychological distance between virtual workers influence their job performance?

To answer these questions, drawing on construal level theory, this study conceptualizes psychological distance and then develops and tests a nomological network to explain the relationship between objective and subjective distance and job performance. By precisely disentangling the effects of different dimensions of objective distance on psychological distance, this study contributes to a more nuanced understanding of how virtual workers construct their perceptions of psychological distance, how it can be measured, and how it influences individual job performance in a VT.

2 Literature Review and Theory Development

2.1 Construal Level Theory

Construal level theory (CLT) explains how psychological distance influences the extent to which an individual’s interpretation about objects or events is abstract or concrete [31, 32]. The general idea of CLT is that the more distant a target is from the individual, the more abstractly and less concretely it will be thought of, focusing more on abstract, holistic, and decontextualized features rather than specific details of information [17]. For example, a distant co-worker may be described as “Annie is clever” (more abstract), whereas a close one may be described as “Annie is able to spot any mistakes in our spreadsheets” (more concrete). Applying construal level theory, VT researchers posit that virtual workers’ psychological distance perceptions determine their construal levels, which in turn affect their judgment and behavior, and enable or constrain their work outcomes. CLT thus challenges the conventional notion that objective distance causes difficulties in sharing a local context and developing a transactive memory (i.e., who knows what) [7]. CLT suggests that even if distant team members were provided with the contextual details normally associated with physical proximity, their subjective perception of the distance would still hinder them from focusing on the provided contextual detail, and thus they would still suffer from the same distance-related difficulties [31]. In other words, virtual workers difficulties, including knowledge sharing, can be explained based on the individual members’ subjective perceptions about the distance between them (herein, psychological distance). In this sense, it has been argued that psychological distance is a direct factor that elicits high levels of construals that disrupt and prevents information flow between virtual workers [36].

2.2 Psychological Distance

Psychological distance refers to “an individual’s perception of how close or how far another person is” [35, p. 983]. Tope and Liberman [32] define it as “a subjective experience that something is close to or far away from the self, here and now” (p. 440). In the VT literature, psychological distance is conceptualized as the extent to which a virtual worker feels he or she is distant from other VT members. Although objective distance and psychological distance are associated, researchers have acknowledged that the role of perceived psychological distance, which influences people’s thoughts, feelings, and behavior, is independent of objective distance [33]. Accordingly, assessing a virtual worker’s psychological distance is important for predicting his or her work outcomes.

Nevertheless, as shown in Table 1, the operationalization of psychological distance is equivocal, which may cause inconsistent findings regarding the effects of psychological distance on job performance. Psychological distance was found to reduce individuals’ willingness to collaborate in the future and to lower their beliefs about the efficacy of distance work [35]. In contrast, psychological distance also affected job performance positively, by increasing creativity and making people focus on essential task features [36]. Some studies found that moderate levels of psychological distance promote trust in information received from others [31]. The present lack of consensus and consistency between definitions necessitates the development and validation of an instrument that precisely captures psychological distance.

Table 1. Operational definitions of psychological distance
Full size table

3 Research Model and Hypotheses

Drawing on CLT, we developed a model to explain how different dimensions of objective distance (spatial, temporal, and cultural) influence a virtual worker’s psychological distance to other VT members and how this distance influences job performance. We outline the relations between objective distance, psychological distance, and job performance, as shown in Fig. 1.

Fig. 1.
figure 1

Research model

Full size image

3.1 Relationship between Objective and Psychological Distance

Objective distance has been widely considered as the most critical and important feature of VTs [6, 25]. Spatial distance refers to the extent to which a virtual worker is dispersed spatially from other members (e.g., different cities or countries), whereas temporal distance refers to the extent to which a virtual worker works in a different time zone from collaborators [10, 25]. Cultural distance is considered important in objective distance. In this study, cultural distance is operationalized as the extent to which a virtual worker works with other members of different nationalities [27]. Cultural distance has great potential to hinder effective interaction within a VT [1, 27].

Each dimension of objective distance can create psychological distance [7, 36], which may cause challenges in judgment and behavior within VTs and become a source of potential conflict. Fujita et al. [12] has recently suggested that because psychologically near events tend to be represented concretely and psychologically distant events tend to be represented abstractly, psychological distance should impede the processing of concrete event representations and facilitate the processing of abstract event representations. Hence, drawing on prior literature, we posit that each dimension of objective distance will influence psychological distance. Apart from a simple linear relationship between objective distance and psychological distance, we hypothesize that interaction effects exist between the dimensions of objective distance. This means that the combination of different objective distance dimensions may increase psychological distance in a non-additive way. Specifically, spatial distance can be overridden by synchronous communication technologies to a certain extent, but temporal distance is difficult to overcome [11, 30]. In particular, lags in feedback and asynchronous communication amplify the problems of mutual knowledge and common ground from which to work and share concerns [11], and these seriously reduce individuals’ perceptions of copresence [13]. Researchers have also suggested that cultural differences interact with spatial distance to produce greater psychological distance than might be expected by the simple combination of the two dimensions of the objective distance [8].

Hypothesis 1:

The different dimensions of objective distance interact in non-additive ways to increase psychological distance.

[H1a]: Temporal distance will positively moderate the relationship between spatial distance and psychological distance

[H1b]: Temporal distance will positively moderate the relationship between cultural distance and psychological distance

[H1c]: Cultural distance will positively moderate the relationship between spatial distance and psychological distance

[H1d]: Cultural distance will positively moderate the relationship between spatial and temporal distances and psychological distance

3.2 Psychological Distance and Job Performance

Construal-level theory [31] explains that people who perceive a high level of psychological distance from their remote partners are more likely to focus on high-level construals (the abstract and decontextualized features of objects) than on the low-level, detailed, and contextualized features of their remote team members’ knowledge. Virtual workers thus tend to fail to share information about the local context, or if they share, other members may ignore the additional context. Studies have suggested that the failure of knowledge contextualization contributes to less effective knowledge utilization [7, 21], which negatively influences a virtual worker’s job performance. Psychological distance also reduces psychological engagement [20]. People who are less psychologically engaged in their roles tend to show passive attitudes toward task performance.

Hypothesis 2:

Psychological distance will be negatively associated with job performance.

4 Methods

4.1 Data Collection

The proposed research model was tested with the collaboration of three multinational companies operating VTs across countries around the world. Executives of the target companies provided support for this research. With this support, we selected VTs where members worked separated by space (e.g., in different cities or countries), time (e.g., different time zones), or were of nationalities (to reflect cultural distance), thus meeting conventional definitions of a VT and providing variation across all dimensions of objective distance. Empirical data was collected by a series of surveys from 163 virtual workers employed across the VTs. The demographics of the respondents are summarized in Table 2.

Table 2. Demographic information of respondents
Full size table

4.2 Measurement

Since the unit of analysis of this study was the individual, all of the research variables were measured at the individual level. We collected team rosters and information about every VT member’s workplace, and nationality. Using this information, we measured the extent to which each virtual worker was dispersed in relation to others spatially, temporally, and culturally. A virtual workers’ spatial distance was measured using the absolute number of air miles with other team members based on Hansen and Lovas [15]. Temporal distance was measured using time zone differences with other members following O’Leary and Cummings [25]. To compute cultural distance, we used nationality differences with other team members by drawing on Raab’s research [27]. Job performance was measured by adapting items used by Mehra et al. [26]. We controlled for gender, age, and organizational tenure.

5 Analysis and Results

5.1 Measurement Model

An explorative factor analysis (EFA) was conducted to evaluate the construct validity of psychological distance. A principal axis factor analysis using the Direct Oblimin method was used for the test. We included measurement items proposed from previous studies, including Hoegl et al. [18], Wilson et al., [35], Lim et al. [23], and Siebdrat et al. [28] into the EFA. After we removed items with low factor loadings, the results of EFA revealed that psychological distance consisted of three sub-dimensions (responsiveness, subjective proximity, and accessibility). Table 3 shows the valid measurement items and the factor loadings corresponding to each item. Next, a confirmatory factor analysis (CFA) was used to validate the instrument. The standardized residuals between the individual instrument items ranged between −0.21 and 0.22, which was well with the recommended bounds for data quality.

Table 3. EFA measurement model results
Full size table

The measurement model was assessed by examining its convergent and discriminant validity. To assess convergent validity, the composite reliability (CR) of each multi-item construct was examined. As shown in Tables 4, all CR values rang from 0.821 to 0.862. To assess discriminant validity, we examined the values for average variance extracted (AVE). For satisfactory discriminant validity, the AVE from a construct should be greater than the variance shared between the construct and other constructs in the model. For each construct, the square root of the average variance extracted should exceed the construct’s correlation with every other construct. The results indicates that the AVE for each construct is larger than the correlation of that construct with all the other constructs in the both models, thus ensuring discriminant validity of the constructs (see Table 4). To ensure that multicollinearity does not pose a problem, we checked values for the index of variance inflation factor (VIF). Our results ranged from 1.00 to 2.120, well below the threshold value of 10, recommended by Harter et al. [16]. The results indicate that it is unlikely that multicollinearity poses a serious concern for this study.

Table 4. Discriminant validity and correlation of constructs
Full size table

Given the structure of the model, the statistical analysis is expected to employ hierarchical linear regression [24] to take into account the different impacts of different sets of independent variables. In particular, we enter the interaction terms of different dimensions of objective distance into the model and regress it to psychological distance in order to verify their non-additive effects. Table 5 represents the results of the model test. Model 1 examines the effects of the control variables on psychological distance. The results indicate that organizational tenure has a negative influence on psychological distance. Three dimensions of objective distance were entered into Model 2. The results show that spatial distance significantly positively influences psychological distance, whereas temporal and cultural distances have no significant influence on psychological distance. Interaction terms between different dimensions of object distance were entered as independent variables into Model 3. The results show that significant interaction effects exist between spatial and cultural distance; spatial and temporal distance; and spatial, temporal, and cultural distance, which account for 12 % of the variance of psychological distance.

Table 5. The results of hierarchical regression analysis
Full size table

6 Discussion

The results showed that psychological distance was the main determinant of job performance and that the different dimensions of objective distance increased psychological distance in a nonlinear fashion. The results showed that interaction effects exist between each dimension of objective distance. That is, the different objective distances increased the psychological distance in a non-additive way. We found that neither cultural nor temporal distance by themselves increased psychological distance but cultural and temporal distance interacted with spatial distance to produce greater psychological distance than did the simple combination of the two dimensions of the objective distance.

Two questions arise. Why is the relationship non-linear, and why is the interaction of objective distance factors more salient than the factors themselves? With increased distance (of any kind) comes a loss of contextual detail, which may hinder collaboration unless it is compensated for. Increases in distance may not produce linear, but instead discrete losses in capability, such as cultural difference becoming too great to understand others’ verbal language unambiguously. Bodenhausen [4] found that making judgments at non-optimal times of the day according to people’s circadian cycle led to stereotypical and thus disproportionally poor judgments.

We assume that some such distance-based collaboration hindrances can be compensated for by other means. Just as individuals possess cognitive reserves by which they can overcome some brain pathologies [e.g., 29] and still function properly, we may think of people’s collaborative reserves, which allow them to overcome collaboration hindrances. For example, a misunderstanding of verbal language may be compensated for by the interpretation of body language. Hence, VT workers may compensate for some levels of cultural distance, but if it interacts with temporal distance (e.g., time zone difference resulting in tiredness and thus poor judgment), then the interaction of both effects may become too much to compensate for, thus leading to a highly salient degradation of collaboration effectiveness.

6.1 Theoretical Implications

Technology-enabled and physically dispersed VTs have emerged as a critical organizational form for structuring work on a global scale [13]. Despite the great potential benefits of VTs, many virtual workers report negative experiences, such as communication depersonalization, conflict, feelings of isolation, and lack of shared knowledge, all of which may negatively influence the virtual worker’s job performance [14]. To address these challenges, research has focused on the effects of objective distance on work outcomes. However, the conflict surrounding the paradox of the “far-but-close” experience for VTs is yet to be reconciled.

By providing a consistent and solid foundation for diagnosis and assessment of psychological distance, this study contributes to VT literature. The findings of this research significantly shift academic and practical attention from objective distance toward subjective distance. A more nuanced understanding of how psychological distance is formed by the different dimensions of objective distance and how psychological distance influences job performance is of academic value.

6.2 Practical Implications

The overarching practical implication of this study is that virtual workers’ psychological distance should be carefully monitored and managed to improve their job performance. Our measurement for objective and psychological distance helps VT managers predict more precisely potential challenges a virtual worker may experience. Given that each VT member has varying degrees of objective distance, VT managers need to pay greater attention to those who have high degrees of distance in its two or more sub-dimensions. VT managers should also carefully monitor virtual workers’ psychological distance at the team level when the team is highly dispersed. Furthermore, our findings imply that VT members’ psychological distance can be mitigated by altering virtual workers’ perceptions of responsiveness, subjective proximity, and accessibility. Technological affordances that can promote emotional and social interaction (e.g., interactivity support, self-presentation, identity expression) may help to mitigate psychological distance.