Perceived organizational support as a mediator between relational exchange and organizational identification

Abstract

Using cross-level data from 364 supervisor–subordinate dyads, we examined how relational exchange quality, perceived organizational support (POS), and organizational identification interrelate. We found subordinate POS mediates the relationship between leader-member exchange (i.e., LMX) and organizational identification. We also found the relational context matters—namely, the immediate supervisor’s relationship with his or her manager (i.e., leader–leader exchange, LLX). Our findings suggest higher quality LLX creates a spillover of resources and reduces the negative association between lower quality LMX and POS. Our study extends both social exchange and social identity theories. First, we delineate how relational exchange quality associates with one’s identity in the organization—placing POS as an integrative mechanism between exchange and identity. Second, we expand the purview of social exchange theory by including other proximal (and interpersonal) relationships as context for social exchange between the individual and organization. Limitations, future research directions, and practical implications are also discussed.

Introduction

Social exchange relationships make up the very fabric of one’s organizational life (Blau, 1964, Gouldner, 1960, Katz and Kahn, 1978). Although a myriad of potential exchange relationships exist within and between organizations, all employees have two seemingly preeminent “relationships at work: one with … [the] immediate supervisor, and one with … [the] organization” (Masterson, Lewis, Goldman, & Taylor, 2000, p. 740). Scholars have provided valuable insights regarding these two social exchange relationships—most commonly represented by leader-member exchange quality (i.e., LMX) and perceived organizational support (i.e., POS). Scholars have generally substantiated the positive association between LMX and POS while also affirming their conceptual and empirical distinction (Masterson et al., 2000, Wayne et al., 2002, Wayne et al., 1997). The quality of social exchange is associated with a myriad of individual-level outcomes such as job satisfaction and performance (for an excellent review see Cropanzano & Mitchell, 2005).

Nevertheless, research on social exchange relationships does not take into account another vital component of organizational life—namely an individual’s sense of belonging and identity. Individuals possess a general and pervasive need for belonging and identity (Baumeister & Leary, 1995). Indeed, building upon social identity theory (Tajfel & Turner, 1986), organizational identification (i.e., “the perception of oneness” with the organization; Ashforth & Mael, 1989, p. 21) provides a backdrop for understanding how identity guides individual behavior and cognition within organizations. Organizational identification, as a result, has been found to be associated with job satisfaction, job involvement, turnover intentions, and in-role and extra-role performance (see meta-analyses, Riketta, 2005). In sum, both social exchange and identity greatly influence one’s organizational experience.

However, these separate research streams beg the question: how do exchange and identity interrelate? Scholars have explored this question within similar streams of research—namely organizational commitment and organizational identification (e.g., Riketta and van Dick, 2005, van Dick et al., 2006, van Knippenberg and Sleebos, 2006). This dilemma is no less pervasive in the investigation of social exchange relationships and organizational identification—only understudied. Theoretical work suggests that positive exchanges will not only increase one’s propensity to reciprocate effort (due to a fair and equitable balance of contributions and benefits) but also increase one’s self-worth and enhance one’s identity (Eisenberger et al., 1986, Graen and Uhl-Bien, 1995). Social exchange relationships, nevertheless, have not been explicitly and consistently shown to associate positively with organizational identification (cf. van Knippenberg & Sleebos, 2006). We propose perceived organizational support (i.e., POS) as an integral mediating mechanism between exchange and identity. Specifically, we argue that POS integrates both the employee’s feelings of fairness (per exchange processes) and feelings of self-worth and self-enhancement (per identity processes). To this end, we shed light on the intersection between exchange and identity by exploring how exchange-based interactions associate with identity-based attitudes.

Our focus on perceived organizational support as a mediator also answers calls from both the ‘relationships at work’ and social exchange literatures. First, ‘relationships at work’ scholars have called to investigate how relationships associate with important outcomes (Dutton and Ragins, 2007, Gersick et al., 2000). Specifically, these scholars urge researchers to explore how experiences at the relational level translate to experiences at the organizational level. Second, social exchange research has consistently found LMX to be related to a multitude of organizationally-focused outcomes yet is silent on how the association occurs (i.e., mediating mechanisms) (Gerstner and Day, 1997, Ilies et al., 2007). In sum, we squarely place POS as an integral mediator between exchange and identity as well as relational and organizational processes—addressing ‘relationships at work’ and exchange literatures.

Finally, scholars have promoted exploring contextual variables that may moderate exchange processes (Hogg et al., 2005, Wayne et al., 2002). Indeed, a major “limitation of LMX theory is that it focuses on dyadic leader-member relationships and fails to consider the fact that these relationships occur in a wider social context” (Hogg et al., 2005, p. 992). To address this call, we focus on a contextual variable that may moderate the association between relational exchange quality and POS. In this study, we focus on the relational context in which the leader-member exchange relationship resides—namely, the immediate supervisor’s relationship with his or her ‘up-line’ manager (i.e., leader–leader exchange [LLX], e.g., Graen et al., 1972, Tangirala et al., 2007). We investigate how LLX moderates the association between LMX and POS. Our study extends social exchange by in that exchange relationships are not created nor maintained in a vacuum. Indeed, social relationships are tied to the social milieu in which they ‘live’ and other proximal relationships are components to this relational context (Hinde, 1997).

Social exchange theory, as alluded to above, provides a major explanatory foundation for LMX, LLX, and POS. Generally, high-quality exchange results when two parties ‘take care of each other’ by exchanging and reciprocating favors (Blau, 1964, Emerson, 1976). The exchange of favors creates “diffuse future obligations”—decreasing the likelihood of keeping an exact tally of favors and increasing the likelihood of engendering a trusting and mutually committed relationship (Blau, 1964, p. 93). These favors, within an organizational setting, include both tangible resources such as monetary rewards, promotions, opportunities for training as well as intangible resources such as informal standing, goal alignment, encouragement and support (Cropanzano & Mitchell, 2005). A positive exchange of resources results in generalized reciprocity. Generalized reciprocity reduces the need for both the equivalence and immediacy of favor return and focuses interests on mutual goals rather than self-centered goals (Gouldner, 1960). Both LMX and POS follow this basic process—albeit on different targets.

Social exchange theory argues that when the leader provides resources in a way that is perceived to be beneficial and equitable, the member will view the relationship positively and reciprocate via increased effort and commitment—resulting in a high quality relationship (Graen & Uhl-Bien, 1995). High quality LMX (or LLX) is assumed to provide appropriate and important resources for the subordinate. High-quality LMX (or LLX) therefore significantly increases the “perceived value of the tangible and intangible commodities exchanged” within the leader-member (or leader–leader) relationship (Wayne et al., 1997, p. 84, emphasis added).

POS scholars, in similar fashion, argue that when the organization provides needed support and resources, the subordinate, in turn, will reciprocate via commitment and effort (e.g., Eisenberger et al., 1986, Masterson et al., 2000). Subordinates perceive support to be particularly high when the organization provides appropriate resources such as rewards, developmental opportunities, and the like (Wayne et al., 2002). As such, POS is defined as the subordinate’s perception of the extent that “their work organization values their contribution and cares about their well-being” due to positive resource allocation (Rhoades & Eisenberger, 2002, p. 698).

Hence, relational exchange quality and POS are distinct yet interrelated constructs. The question is: how are they interrelated? Resources (e.g., rewards, developmental opportunities, feedback) are usually implemented within (or in the least, through) the supervisory relationship. Supervisors control many of the organizational resources made available to employees and are, therefore, perceived as organizational agents (cf. Chen, Tsui, & Farh, 2002). Supervisors frequently interact with subordinates with regard to allocating resources (e.g., merit increases, performance incentives, training, feedback)—providing fertile ground for the exchange of organizationally-derived resources. Given that high-quality LMX provides an increased amount of resources and the leader is perceived as an organizational agent, the subordinate will perceive support from the organization. Hence, we hypothesize that:

H1

Subordinate LMX will be positively related to subordinate POS

Immediate supervisors have sway over a broad range of organizationally-derived resources integral in the POS process. However, exchange relationships are nested within salient hierarchical relationships. We argue that the leader–leader relationship (i.e., LLX) forms an important part of a relational context that surrounds the subordinate’s organizational experience—especially with regard to POS (Hinde, 1997, Hogg et al., 2005). LLX assesses the quality of the ‘up-line’ manager-immediate supervisor (i.e., leader–leader) relationship. Similar to LMX, higher quality LLX relationships provide increased tangible and intangible resources for the immediate supervisor. However, the immediate supervisor is able to re-distribute these resources (granted within the leader–leader relationship) to his or her subordinates. As such, the leader–leader relationship, via LLX, provides a base pool of tangible and intangible resources that the immediate supervisor can then re-distribute to the subordinate. The immediate supervisor distributes pieces of the resource pie to the subordinate, but the ‘up-line’ manager plays a role in deciding how big that pie is.

Following our arguments above, higher LLX quality will result in more tangible and intangible resources available for distribution to ‘front-line’ subordinates whereas lower LLX quality will result in fewer. Traditional LMX theory supposes that supervisors distribute more resources to the in-group (i.e., those subordinates in higher quality LMX relationships) and fewer resources to the out-group (i.e., those subordinates in lower quality LMX relationships). This is very much the case in the condition of lower quality LLX. In this condition, the immediate supervisor will distribute the scarce resources to those of the subordinate in-group (i.e., higher LMX) more readily and frequently than to those of the subordinate out-group (i.e., lower LMX). However, in the case of higher quality LLX, the immediate supervisor receives increased resources resulting in more that are available for re-distribution. We argue that these increased resources will tend to spillover to more subordinates—even those experiencing a lower quality relationship with the immediate supervisor (van Dijk & de Cremer, 2006). ‘Lower quality’ subordinates in the highLLX condition will receive more resources and, therefore, perceive higher organizational support than the ‘lower quality’ subordinates in the low LLX condition. To be sure, in the high LLX condition, supervisors will still provide significant resources to ‘higher quality’ subordinates maintaining levels of organizational support. The difference lies in that those ‘lower quality’ subordinates will also experience increased resources and thus perceive increased support. In sum, increased available resources (again, due to the high LLX condition) will spillover to those subordinates with lower LMX quality—thus weakening the association between LMX and POS. Hence, we hypothesize:

H2

The relationship between subordinate LMX and subordinate POS is moderated by LLX such that the relationship is weaker when LLX is higher than when it is lower.

Scholars have found that POS increases employees’ feelings of obligation and positive reciprocity (e.g., Rhoades & Eisenberger, 2002). However, organizational support may not only increase reciprocity but also feelings of self-enhancement (i.e., higher self-esteem due to organizational membership; Ashforth & Mael, 1989). To be sure, organizational membership that increases one’s feelings of self-worth and self-esteem will stimulate organizational identification (i.e., including the organization in one’s self-concept; Ashforth & Mael, 1989). Organizational support affirms the subordinate’s value and informal standing as well as increases the organization’s perceived attractiveness—all important amplifiers of self-enhancement and organizational identification (i.e., moving from situated identification to deep-structure identification; Riketta, van Dick, & Rousseau, 2006). To point, perceptions of favorable standing within the organization have been found to increase organizationally-based self-esteem and positive mood (e.g., Lee & Peccei, 2007). Thus, we hypothesize that:

H3

Subordinate’s POS will be positively related to the subordinate’s organizational identification.

Feeling supported and valued by the organization (i.e., POS) glues positive (and proximal) work experiences (such as high quality relationships) with positive (yet distal) behaviors and attitudes that benefit the organization. Indeed, POS has been found to mediate various relational processes such as turnover and organizational politics (Allen et al., 2003, Hochwater et al., 2003). Specifically, these two studies converge on the basic logic that positive ‘subordinate experiences’ (e.g., supportive human resource practices, few political maneuverings) engender perceptions of organizational support which, in turn, increase behaviors and attitudes that reciprocate that support. We expand this logic by arguing that these proximal subordinate experiences (as mentioned above) are primarily relational in nature and can be captured via LMX quality. Indeed, Hochwater et al. (2003) delineated political maneuverings across hierarchical dyads and Allen et al. (2003) operationalized supportive human resource practices as participation in decision making. Similarly, we argue a high LMX with the supervisor (as organizational agent) will signal that the organization values the subordinate. The felt organizational support and ‘valuing,’ in turn, provides increased feelings of self-worth. Feelings of self-worth and esteem tend to increase self-enhancement—wherein the individual’s identity is enhanced (Edwards, in press). Thus, we argue that self-enhancement, due to POS, will increase organizational identification. Hence:

H4

Subordinate POS will mediate the relationship between subordinate LMX and organizational identification.