Scale Indicators of Social Exchange Relationships:A Comparison of Relative Content Validity
Jason A. Colquitt and Michael D. BaerUniversity of Georgia
David M. LongCollege of William and Mary
Marie D. K. Halvorsen-GanepolaUniversity of Notre Dame
Although social exchange theory has become one of the most oft-evoked theories in industrial andorganizational psychology, there remains no consensus about how to measure its key mechanism: socialexchange relationships (Blau, 1964). Drawing on Cropanzano and Byrne’s (2000) review of contempo-rary social exchange theorizing, we examined the content validity of perceived support, exchange quality,affective commitment, trust, and psychological contract fulfillment as indicators of social exchangerelationships. We used Hinkin and Tracey’s (1999) quantitative approach to content validation, whichasks participants to rate the correspondence between scale items and definitions of intended (andunintended) constructs. Our results revealed that some of the most frequently utilized indicators of socialexchange relationships—perceived support and exchange quality—were significantly less content validthan rarely used options like affect-based trust. Our results also revealed that 2 direct measures—Bernerth, Armenakis, Feild, Giles, and Walker’s (2007) scale and a scale created for this study—werecontent valid. We discuss the implications of these results for future applications of social exchangetheory.
Keywords: social exchange, trust, citizenship
Why might fair treatment cause employees to attend optionalmeetings? How could reductions in training lead to increases intheft? Why might getting a “special deal” at work result in higherlevels of task performance? Such questions have, increasingly,come to be examined through the lens of social exchange theory,which explains how different types of benefits are exchangedaccording to various rules and how such exchanges foster high-quality relationships (see Cropanzano & Mitchell, 2005, for areview). Indeed, as Table 1 reveals, social exchange theory hasbecome one of the most oft-evoked theories in the Journal ofApplied Psychology. Its propositions have been used to explainphenomena in a variety of literatures, including justice (Cropan-zano & Rupp, 2008), employment relationships (Coyle-Shapiro &Conway, 2004), mentoring (Eby et al., 2013), citizenship behavior(Organ, 1988, 1990), and counterproductive behavior (Greenberg& Scott, 1996), to name a few.Given such ubiquity, one might assume that social exchange
theory excels on the two criteria often used to evaluate theories:
utility and falsifiability (Bacharach, 1989). Unfortunately, wewould argue, such an assumption is only half-right, at least interms of the way the theory is applied in industrial and organiza-tional psychology. It is not clear that empirical refutation of thetheory’s propositions is possible, because there is no consensus onhow to measure its core mechanism: social exchange relationships(Blau, 1964). Reviews of the theory have pointed to a number ofpotential indicators of social exchange relationships, includingperceived support, exchange quality, affective commitment, trust,and psychological contract fulfillment (Cropanzano & Byrne,2000; Cropanzano & Mitchell, 2005; Cropanzano & Rupp, 2008;Cropanzano, Rupp, Mohler, & Schminke, 2001). If some of thosevariables are less construct-valid indicators of social exchange rela-tionships than others, tests of the theory’s propositions become diffi-cult to interpret (Bacharach, 1989). Moreover, if some of thoseindicators actually occupy functionally distinct and different posi-tions in the theory’s causal chain, tests using those indicators willmuddy the logical adequacy of the theory, harming falsifiability(Bacharach, 1989).Our purpose in this study was to provide a critical examination
of the relative validity of the social exchange relationship indica-tors used in the literature. To do so, we applied Hinkin andTracey’s (1999) quantitative approach to content validation, withcontent validity defined as the extent to which a measure’s itemsreflect a theoretical content domain (Kerlinger & Lee, 2000;Schwab, 1980). Hinkin and Tracey’s approach asks respondents toexplicitly rate the correspondence of scale items to a construct’sstated definition. We used this method to gauge the correspon-dence of measures of perceived support, exchange quality, affec-
This article was published Online First April 7, 2014.Jason A. Colquitt and Michael D. Baer, Department of Management,
University of Georgia; David M. Long, Department of OrganizationalBehavior, College of William and Mary; Marie D. K. Halvorsen-Ganepola,Department of Management, University of Notre Dame.Correspondence concerning this article should be addressed to Jason A.
Colquitt, University of Georgia, Department of Management, Terry Col-lege of Business, 412 Brooks Hall, Athens, GA 30602-6256. E-mail:[email protected]
tive commitment, trust, and psychological contract fulfillment tothe conceptual definition of social exchange relationships (Blau,1964). We also used this approach to examine the content validityof scales that were explicitly created to assess social exchangerelationships (e.g., Bernerth, Armenakis, Feild, Giles, & Walker,2007; Shore, Tetrick, Lynch, & Barksdale, 2006), including onedeveloped for this study. Our results reveal a number of surprises,with some of the most oft-used operationalizations possessingsome potentially important content validity problems.
Social Exchange: The Theory and the RelationshipsAs summarized in reviews of the literature, “social exchange
theory” is actually a multidisciplinary family of perspectives ratherthan one single model (Cropanzano & Byrne, 2000; Cropanzano &Mitchell, 2005; Cropanzano & Rupp, 2008; Cropanzano et al.,2001). We focus on what is sometimes called “contemporarysocial exchange theory,” which can be traced to Organ’s (1988,1990) application of Blau’s (1964) writings on social exchangerelationships (Cropanzano et al., 2001). Blau (1964) describedrelationships, or “social associations,” as “an exchange of activity,tangible or intangible, and more or less rewarding or costly,between at least two persons” (p. 88). He went on to distinguishtwo specific kinds of relationships. Social exchanges represent amore invested relationship that is based on—and motivated by—obligatory exchanges of unspecified favors and benefits, over anopen-ended and long-term time frame. In contrast, economic ex-changes represent a less invested and more contractual relationshipwhere benefits and repayment schedules are clearly specified.Blau’s (1964) discussion of social exchange relationships men-
tions a number of benefits, defined here as voluntary, beneficialactions by one exchange partner that are expected to create a desireto give back on the part of the other. Those include assistance,advice, compliance, appreciation, and instrumental services. Forthe most part, such benefits are symbolic and particularistic, mean-ing that the identity of the provider impacts the value of the benefit(Foa & Foa, 1980). Of course, concepts such as assistance andcompliance can be both quid and quo. If one exchange partnerprovides assistance to the other, subsequent acts of compliance bythat other could constitute reciprocative behavior, defined here as
voluntary, beneficial actions by one exchange partner that arebelieved to exemplify giving back to the other.Organ (1988, 1990) applied Blau’s (1964) theorizing to citizen-
ship behavior in arguing that justice on the part of an organizationcould explain instances of an employee being a “good soldier.”From this perspective, justice acts as a benefit that is positivelyassociated with a social exchange relationship, with that relation-ship being positively associated with the reciprocative behavior ofcitizenship. Organ’s work molded contemporary social exchangetheorizing in two ways. First, he expanded Blau’s (1964) concep-tualization of exchange relationships to include person–organization linkages, putting organizations in the role of juristicpersons (Folger & Cropanzano, 2001) who could provide benefitsand receive reciprocation. Second, his benefit ¡ social exchangerelationship ¡ reciprocative behavior causal chain has come toform the spine of subsequent empirical tests (Cropanzano et al.,2001). Some of those tests focused on organizational citizenshipbehavior targeted at a supervisor (Masterson, Lewis, Goldman, &Taylor, 2000), termed OCBS (Malatesta & Byrne, 1997). Othertests focused on citizenship behavior targeted at an organization(e.g., Konovsky & Pugh, 1994; Masterson et al., 2000; Moorman,Blakely, & Niehoff, 1998), termed OCBO (Williams & Anderson,1991).
From Conceptual to Operational: Measuring SocialExchange Relationships
Notably, the empirical studies that occurred in the wake ofOrgan’s (1988, 1990) theorizing used different variables to capturethe mediating role of social exchange relationships. Organ (1988,1990) himself did not suggest a mediator to capture Blau’s (1964)dynamic, focusing his energies on defining the still nascent citi-zenship behavior construct. Moreover, industrial and organiza-tional psychology seemed to already possess off-the-shelf vari-ables that reflected elements of Blau’s (1964) theorizing.Cropanzano and Byrne (2000) provided the first discussion of thismediational dissensus. Noting that an operationalization of socialexchange relationships was the “missing linchpin” in contempo-rary theorizing, they reviewed five “candidates” that could capturethat mediator (Cropanzano & Byrne, 2000, pp. 150–151): per-
Table 1Frequency of Citations to Theories in the Journal of Applied Psychology (2000–2010)
Theory nameExact phrase used inGoogle Scholar search
Citations from2000 to 2010
Theory of planned behavior Ajzen, 1991 136Social cognitive theory Bandura, 1977 112Social exchange theory Blau, 1964 106Social identity theory Tajfel & Turner, 1986 80Expectancy theory Vroom, 1964 79Goal-setting theory Locke & Latham, 1990 78Equity theory Adams, 1965 71Self-determination theory Deci & Ryan, 1985 60Affective events theory Weiss & Cropanzano, 1996 53Job characteristics theory Hackman & Oldham, 1976 33
Note. We generated the list of theories using the references in S. W. J. Kozlowski’s (2012) Oxford Handbookof Organizational Psychology. The citation search was performed with Google Scholar’s “with the exactphrase,” “Return articles published in,” and “Return articles dated between” fields.
ceived support, exchange quality, affective commitment, trust, andpsychological contract fulfillment. As shown in Figure 1, thesevariables have gone on to become the mediators of choice forscholars wanting to explain the linkage between some benefit andsome reciprocative behavior.Unfortunately, as reviewed in the sections below, many of these
variables possess content validity problems when used as opera-tionalizations of social exchange relationships (see Appendix A fora listing of the scale items). In some cases the items do notsufficiently evoke a relationship, focusing instead on statementsabout one exchange partner. In other cases the items do notsufficiently evoke the sentiments that Blau (1964) ascribed tosocial exchange relationships. For example, Blau (1964) noted thatsocial exchange relationships entail “unspecified obligations” (p.93), require “trusting others” (p. 94), constitute “commitments tothe other party” (p. 98), and possess “elements of intrinsic signif-icance” (p. 112). In still other cases, the items explicitly ask aboutbenefits or reciprocative behaviors, injecting contaminated contentthat belongs to antecedents or consequences of social exchangerelationships.
Perceived SupportPerceived support is defined as the degree to which one ex-
change partner values the contributions of the other and showsconcern for his or her well-being (Eisenberger, Huntington,Hutchison, & Sowa, 1986). When targeted at an organization, theconstruct is labeled perceived organizational support, or POS(Eisenberger et al., 1986). When targeted at a supervisor, theconstruct is labeled perceived supervisor support, or PSS (Eisen-berger, Stinglhamber, Vandenberghe, Sucharski, & Rhoades,2002). Perceived support represents the single most commonmeans of capturing the mediating role of social exchange relation-ships, with the benefits examined including justice, developmentalexperiences, promotions, and inclusion and the reciprocative be-haviors including citizenship behavior, counterproductive behav-ior, job performance, and turnover (e.g., Allen, Shore, & Griffeth,2003; El Akremi, Vandenberghe, & Camerman, 2010; Mastersonet al., 2000; Moorman et al., 1998; Tekleab, Takeuchi, & Taylor,2005; Wayne, Shore, Bommer, & Tetrick, 2002; Wayne, Shore, &Liden, 1997).
Num
ber
of A
rtic
les
0
5
10
15
20
!" #" $" %" &" '" (" )" *"PerceivedSupport
ExchangeQuality
AffectiveCommitment
Trust PCF
Figure 1. Frequency with which variables have been used to test the mediating role of social exchangerelationships in top industrial and organizational psychology journals (2000–2010). Journals included theJournal of Applied Psychology, Academy of Management Journal, Personnel Psychology, OrganizationalBehavior and Human Decision Processes, Journal of Management, and Journal of Organizational Behavior.PCF ! psychological contract fulfillment.
Operationally speaking, perceived support is viewed as highwhen supervisors or organizations help employees, value theircontributions, consider their goals and values, and care about theirwell-being. The items do not evoke a relationship per se; nor dothey reflect the sentiments that Blau (1964) ascribed to socialexchange. Instead, they focus solely on benefits provided by anexchange partner, resembling the assistance and appreciation thatBlau used to exemplify benefits. That focus is understandablegiven that the construct was meant to be an indicator of a super-visor or organization’s commitment to an employee (Eisenbergeret al., 1986). Nonetheless, that focus may inject contaminationwhen social exchange relationships are measured. It may alsocreate ambiguity of causal direction when relationships are mod-eled with other benefits. For example, if interpersonal justicereflects the degree to which a supervisor treats an employee withrespect (Colquitt, 2001), how can an interpersonal justice ¡ PSScausal flow be modeled without raising concerns about nonrecur-siveness?
Exchange QualityExchange quality has come to be defined as the degree to
which a relationship between two exchange partners is charac-terized by mutual respect, trust, and obligation (Gerstner &Day, 1997; Graen & Uhl-Bien, 1995). When targeted at asupervisor– employee dyad, exchange quality is termed leader–member exchange, or LMX (Graen & Uhl-Bien, 1995). Whentargeted at an organization–employee dyad, exchange quality istermed organization–member exchange, or OMX (Karriker &Wil-liams, 2009). Exchange quality represents the second most com-mon means of capturing the mediating role of social exchangerelationships, with the benefits in those studies including justice,rewards, and punishment and the reciprocative behaviors in thosestudies including citizenship behavior, learning, job performance,and turnover (e.g., Cropanzano, Prehar, & Chen, 2002; El Akremiet al., 2010; Masterson et al., 2000; Tekleab et al., 2005; Walum-bwa, Cropanzano, & Hartnell, 2009; Wayne et al., 2002).Operationally speaking, the measurement of exchange quality
has evolved over the years. Graen and Uhl-Bien (1995) argued thatthe “centroid item” (p. 236) for assessing exchange quality asksabout the perceived effectiveness of the working relationship withan exchange partner. Other items ask about the beliefs and inten-tions of both the exchange partner and the focal respondent. Thatcentroid item gives exchange quality at least one item that explic-itly focuses on relationships. Although there is an undercurrent oftrust in the items, the scale does not seem to assess sentiments suchas obligation, commitment, or significance. Moreover, many of theitems seem to reference past or future benefits (e.g., information,understanding, assistance) or reciprocation (e.g., defending a part-ner’s actions). Such content could create ambiguity of causaldirection when exchange quality is used as a mediator. For exam-ple, if informational justice reflects the degree to which a super-visor provides adequate explanations to an employee (Colquitt,2001), modeling an informational justice ¡ LMX causal flow isproblematic if one indicator of LMX is a supervisor telling em-ployees where they stand. Similarly, modeling an LMX¡ OCBScausal flow is problematic if another indicator of LMX is anemployee being willing to help defend a supervisor, given thatsuch an action would itself comprise OCBS.
Affective CommitmentAffective commitment reflects the degree to which one ex-
change partner has an emotional attachment to the other, such thatthe partner identifies with and is involved in the shared association(Allen & Meyer, 1990; see also Mowday, Porter, & Steers, 1982).When targeted at an organization, the construct is termed affectiveorganizational commitment. When targeted at a supervisor, theconstruct is termed affective supervisory commitment. Affectivecommitment has been used to capture the mediating role of socialexchange relationships in several studies, with the benefits includ-ing justice, participation, fit, and growth opportunities and thereciprocative behaviors including citizenship behavior, counter-productive behavior, job performance, and turnover (e.g., Allen etal., 2003; Cropanzano, Rupp, & Byrne, 2003; Eisenberger et al.,2010; Lavelle et al., 2009; Liao, Joshi, & Chuang, 2004).Operationally speaking, affective commitment is viewed as high
when employees feel attached to or “part of the family” with theirorganizations or supervisors and when they draw personal mean-ing and a sense of belonging from those linkages. None of theitems mention a relationship explicitly. They do tap some of thesentiments that Blau (1964) ascribed to social exchange relation-ships, such as a bond of commitment and a sense of intrinsicsignificance. Blau (1964) noted that such sentiments were neededto give relationships the stability needed for long-term, open-endedarrangements. Moreover, none of the items seem to referencebenefits or reciprocation, preventing some of the contaminationobserved for perceived support and exchange quality. Thus, theitems used to assess affective commitment seem to be a content-valid reflection of the social exchange relationship domain.
TrustThe scholarly literature on trust is marked by two distinct
definitions. Mayer, Davis, and Schoorman (1995) defined trust asthe willingness of one exchange partner to be vulnerable to theactions of another, based on the expectation that the other willperform some action in the absence of monitoring or control.McAllister (1995) defined trust as positive expectations about anda willingness to act upon the words and intentions of an exchangepartner. Although Mayer et al.’s (1995) willingness to be vulner-able (WBV) conceptualization is unidimensional, the positive ex-pectations described by McAllister (1995) come in two varieties.Cognition-based trust (CBT) is rooted in rational assessments oftrustworthiness. Affect-based trust (ABT), in contrast, is rooted inthe emotional ties that bond exchange partners together. Takentogether, these three conceptualizations of trust have been used tocapture the mediating role of social exchange relationships in ahandful of studies, with the benefits including justice, empower-ment, inducements, and communication and the reciprocative be-haviors including citizenship behavior and job performance (e.g.,Aryee, Budhwar, & Chen, 2002; Colquitt, LePine, Piccolo, Zapata,& Rich, 2012; Huang, Iun, Liu, & Gong, 2010; Konovsky & Pugh,1994; Montes & Irving, 2008; Yang, Mossholder, & Peng, 2009).Operationally speaking, CBT is viewed as high when organiza-
tions or supervisors act in a professional, competent, and dedicatedmanner. For its part, ABT is high when the relationship between anemployee and his or her supervisor or organization is a sharing oneand when it is marked with mutual care, communication, andinvestment. Finally, WBV is high when an employee is willing to
“stick his/her neck out” when dealing with a supervisor or orga-nization. Although trust is clearly central to Blau’s (1964) concep-tualization of social exchange, the three operationalizations seemto differ in their content validity. Of the three trust variants, ABTmost explicitly evokes a relationship and most explicitly evokessentiments like commitment and intrinsic significance. CBT, incontrast, focuses solely on benefits provided by an exchangepartner and was framed as an antecedent of ABT by McAllister(1995). WBV, for its part, does seem to capture the trust senti-ments described by Blau (1964), though the items about speakingfreely and being creative also signify an intention to reciprocate,potentially evoking reciprocative behavior. That positioning isconsistent with McAllister, Lewicki, and Chaturvedi (2006), whoviewed both CBT and ABT as antecedents of WBV.
Psychological Contract FulfillmentA psychological contract reflects an exchange partner’s belief
that certain benefits are promised by an other, in exchange forcertain contributions on his or her part (Rousseau, 1990). Thenature of those promised benefits and contributions has been usedto distinguish transactional contracts from relational contracts. Theformer reflect an exchange of pay and advancement for hard workand advance notice of departure, with the latter reflecting anexchange of job security and development for loyalty and citizen-ship (e.g., Robinson, Kraatz, & Rousseau, 1994; Rousseau, 1990).Psychological contract fulfillment (PCF), in turn, reflects an over-all evaluation of the degree to which an exchange partner hasfulfilled the obligations that were promised (Robinson & Morri-son, 1995, 2000; Robinson & Rousseau, 1994). PCF has been usedto capture the mediating role of social exchange relationships in ahandful of studies, with the benefits including justice, accounts,and (lack of) politics and the reciprocative behaviors includingcitizenship behavior, counterproductive behavior, job perfor-mance, and turnover (e.g., Kickul, Neuman, Parker, & Finkl, 2001;Lester, Kickul, & Bergmann, 2007; Rosen, Chang, Johnson, &Levy, 2009; Tekleab et al., 2005).Operationally speaking, PCF is viewed as high when employees
feel that the promises made by their supervisors or organizationshave been kept. Although there is a dyadic quality to the PCFitems, they do not explicitly refer to the relationship with theexchange partner. They also do not reference social exchangesentiments such as obligation, trust, commitment, or significance.Instead, they seem to focus on supervisor or organizational bene-fits, in the form of promise fulfillment. Given that benefits focus,it is perhaps not surprising that PCF is typically cast as an ante-cedent of affective commitment and trust (Robinson & Morrison,1995; Robinson & Rousseau, 1994; Rosen et al., 2009; Zhao,Wayne, Glibkowski, & Bravo, 2007).
mitment, trust, and PCF have their roots in exchange theorizing tovarying degrees, none of the scales used to measure them werecreated for the express purpose of assessing Blau’s (1964) con-ceptualization of social exchange relationships. Instead, they weretaken off the shelf to meet the demand for mediators created byOrgan’s (1988, 1990) reintroduction of Blau’s (1964) ideas (Cro-
panzano & Byrne, 2000; Cropanzano & Mitchell, 2005; Cropan-zano & Rupp, 2008; Cropanzano et al., 2001). Our study willexamine whether those variables can in fact serve as content-validindicators of social exchange relationships, or whether they arecontaminated by Blau’s (1964) benefit or reciprocative behaviorconcepts. In so doing, we will also examine the content validity ofthree scales that were explicitly created to assess social exchangerelationships.The first of those scales is Shore et al. (2006), which operation-
alizes social exchange relationships as mutual trust and invest-ment, a long time duration, and socioemotional (rather than finan-cial) transactions. Although both relationships and social exchangesentiments are evoked in the items, they also seem to tap benefits(i.e., investment, rewards, care) and reciprocative behaviors (e.g.,effort, hard work, assistance). Nevertheless, Shore et al.’s measurehas been used in recent studies, with the benefits including justice,leadership, and inducements and the reciprocative behaviors in-cluding citizenship and job performance (Hom et al., 2009; Rupp& Cropanzano, 2002; Song, Tsui, & Law, 2009). The second ofthose scales is Bernerth et al. (2007), intended to be a measure ofexchange quality that would more explicitly evoke exchange andreciprocity, relative to more seminal scales (Graen & Uhl-Bien,1995). As with Shore et al.’s scale, some items explicitly evokerelationships while also hinting at the social exchange sentimentsdescribed by Blau (1964). Also like Shore et al. (2006), some itemsseem to tap benefit and reciprocative behavior themes, though in aless defined and specific fashion. We are not aware of a study thathas used Bernerth et al.’s (2007) scale to mediate the relationshipbetween some benefit and some reciprocative behavior. Finally,the third scale, labeled the Social Exchange Relationship Scale(SERS) in Appendix A, was created for the purposes of this study.It explicitly asks whether a given relationship is characterized bythe sentiments that Blau (1964) described in his theorizing: mutualobligation, mutual trust, mutual commitment, and mutual signifi-cance.
Method
Sample
Four hundred undergraduate students at a large southeasternuniversity were recruited for this study from a research poolconsisting of juniors and seniors majoring in business manage-ment. Students received course credit for their participation. Par-ticipants agreed to complete three online surveys that were pre-sented to them at 3-week intervals. Three hundred and thirty-oneparticipants completed all three surveys, representing an overallresponse rate of 83%. Given the importance of careful respondingin survey studies (Berry et al., 1992; Clark, Gironda, & Young,2003; Schmitt & Stults, 1985)—especially those using Hinkin andTracey’s (1999) methodology—we included a number of carelessrespondent checks. These included monitoring the time spent onthe instructions screen, the time spent on the survey as a whole,and the inclusion of an instructed response item that asked partic-ipants to “Please click the circle under __” (Meade & Craig, 2012).The content of the blank was “2” in the first survey, “4” in thesecond, and “1” in the third. These screens resulted in a finalsample size of 234. The participants were 57% male, had anaverage age of 21.2 years (SD ! 2.86), and reported the following
ethnicities: 83% Caucasian, 7% Asian, 6% African American, 3%Hispanic, and 2% Other. Thirty-nine percent of the participantswere employed. Schriesheim, Powers, Scandura, Gardiner, andLankau (1993) argued that undergraduates are an appropriate sam-ple for content validity studies, given that the primary requirementis the intellectual ability to evaluate the consistency between scaleitems and construct definitions. Hinkin and Tracey (1999) echoedthis recommendation when outlining their procedure.
ProcedureGiven that social exchange theory is applied to both supervisor
and organization targets, we included both supervisor and organi-zation conditions in our study. As described by Hinkin and Tracey(1999), the first step in their quantitative approach to contentvalidation is identifying definitions for the constructs of interest.Drawing on Blau’s (1964) descriptions, we used the followingdefinition for social exchange relationships:
Social exchange relationship. “A more invested relationship betweenan employee and his/her [supervisor/organization] that is based on—and motivated by—obligatory exchanges of unspecified favors andbenefits, over an open-ended and long-term time frame.”
Also drawing on Blau’s (1964) descriptions and examples, weused the following definition for the benefits that are believed tofoster social exchange relationships:
Benefit. “A voluntary, beneficial action by [a supervisor/an organiza-tion] that is expected to create a desire to ‘give back’ amongemployees.”
Finally, again drawing on Blau’s (1964) descriptions and exam-ples, we used the following definition for the reciprocative behav-iors believed to be triggered by social exchange relationships:
Reciprocative behavior. “A voluntary, beneficial action by an em-ployee that is believed, by the employee, to exemplify ‘giving back’to [a supervisor/an organization].”
The next step of Hinkin and Tracey’s (1999) procedure involveshaving respondents rate the correspondence between scale itemsand a given construct definition. In our study, participants wererandomly assigned to either the supervisor condition or the orga-nization condition. They then filled out a survey in which theperceived support, exchange quality, affective commitment, trust,PCF, and direct measure items were paired with the social ex-change relationship definition, the benefit definition, or the recip-rocative behavior definition. The definition received at Time 1 wasrandomly selected by the survey software. Three weeks later, thesame items were paired with a different definition, which wasrandomly selected from the remaining two definitions. The sameprocess was repeated with the remaining definition 3 weeks afterthat. The temporal separation of surveys was performed for tworeasons: (a) to minimize fatigue on the part of participants and (b)to reduce the likelihood that responses to one definition would beretained in working memory when responding to other definitions(Podsakoff, MacKenzie, Lee, & Podsakoff, 2003). The ordering ofscales within each of the three surveys was randomized to preventitem context effects (Podsakoff et al., 2003). The survey softwarerandomly presented all scales to the participants with no restric-tions or groupings imposed by the researchers. The order of items
within each scale was not randomized, with items presented intheir published order, as shown in Appendix A.Participants were asked to indicate the extent to which items
were consistent with the definition provided using this 7-pointscale: 1 ! Item does an extremely bad job of measuring theconcept to 7 ! Item does an extremely good job of measuring theconcept. Consider, for example, a participant in the supervisortarget condition who received the benefit definition on the firstsurvey, the social exchange relationship definition on the secondsurvey, and the reciprocative behavior definition on the third survey.That participant would react to, for example, the perceived supportitem, “My supervisor takes pride in my accomplishments,” by firstjudging its correspondence to the benefit definition on the 7-pointscale. Three weeks later the same item, when encountered again onthe survey, would be judged against the social exchange relation-ship definition on the same scale, with that process repeated 3weeks later with the reciprocative behavior definition. Given thepresence of negatively worded items in some scales, the instruc-tions clarified that some items would measure a concept by indi-cating high levels of it, and others would measure the concept byindicating low levels of it.In the end, each participant wound up with three different mean
levels of correspondence for each scale: a mean for the benefitdefinition, a mean for the social exchange relationship definition,and a mean for the reciprocative behavior definition. The final stepof Hinkin and Tracey’s (1999) procedure was testing the statisticalsignificance of the differences between those means. We utilizedrepeated-measures analyses of variance (ANOVA) to first identifywhether there were significant differences among the three meansas a set. We then performed post hoc pairwise comparisons to testthe specific difference between the social exchange relationshipmean and the benefit mean and the specific difference between thesocial exchange relationship mean and the reciprocative behaviormean. These comparisons were performed with the Sidak–Bonferroni adjustment (Sidak, 1967). This adjustment corrects forthe familywise error rate for multiple comparisons but has theadvantage of tempering the Bonferroni adjustment’s adverse im-pact on statistical power (Keppel & Wickens, 2004).
MeasuresAll of our measures are shown in Appendix A. Because each
scale was administered three times, we provide alphas for each ofthe three definitional conditions.Perceived support. We used Eisenberger, Armeli, Rexwin-
kel, Lynch, and Rhoades’s (2001) measure of perceived support.Exchange quality. We used a version of Scandura and
Graen’s (1984) exchange quality scale that was adapted to haveLikert-style anchors instead of idiosyncratic anchors (Lee, Scan-dura, Kim, Joshi, & Lee, 2012; Liden, Wayne, & Stillwell, 1993;Tekleab & Taylor, 2003).Affective commitment. Meyer, Allen, and Smith’s (1993)
measure of affective commitment was utilized.Willingness to be vulnerable. We measured WBV using an
adaptation of Schoorman, Mayer, and Davis’s (2007) trust scale.This scale represents an updated version of the measures intro-duced in Mayer and Davis (1999) and Mayer and Gavin (2005).Affect-based trust. McAllister’s (1995) measure of ABT was
Cognition-based trust. McAllister’s (1995) measure of CBTwas also utilized in our study.Psychological contract fulfillment. We used Robinson and
Morrison’s (2000) measure of PCF.Direct measures. Shore et al.’s (2006) measure of social
exchange was included as a direct measure of the construct, alongwith Bernerth et al.’s (2007) scale. We also included our four-itemSERS, created for this study.
ResultsTable 2 provides our repeated-measures ANOVA results. We
present results for the overall sample and also for the more specificsupervisor and organization conditions, given that some scholarswork exclusively with one target or another. We also provideitem-level results in Appendix A. Given the wider standard errorsfor item-level results, we collapse across targets in order to booststatistical power. In general, the item-level results affirm theresults at the scale level, so we focus on the latter in the discussion
below. Note that Table 2 and Appendix A provide the mean levelof definitional correspondence for the social exchange relationshipcondition and also provide rWG values to illustrate the within-condition agreement for that correspondence (James, Demaree, &Wolf, 1984).The overall F results in Table 2 show that there were significant
differences among a given scale’s mean correspondence for thebenefit, social exchange relationship, and reciprocative behaviordefinitions, for all 10 of the scales included in our study. Theseoverall differences were observed for the complete sample andalso for the supervisor target conditions. In the organization targetcondition, these overall differences were observed for all scalesexcept affective commitment.Given those significant overall effects, we probed the more
specific patterns using the post hoc pairwise comparisons shown inTable 2. The table illustrates the mean correspondence differencewhen a scale was referenced to the social exchange relationshipdefinition, relative to both the benefit definition and the recipro-
Note. n ! 234 for the overall sample; n ! 112 for the supervisor target; n ! 122 for the organization target.Bolded numbers indicate that the scale’s mean definitional correspondence is statistically significantly higher forthe social exchange relationships definition than for both the benefit and reciprocative behavior definitions.Cohen’s d values are in parentheses next to the mean differences. SER! social exchange relationship definition;Ben! benefit definition; Rec! reciprocative behavior definition; WBV! willingness to be vulnerable; ABT!affect-based trust; CBT! cognition-based trust; PCF! psychological contract fulfillment; Shore! Shore et al.(2006) scale assessing social exchange relationships; Bernerth ! Bernerth et al. (2007) scale assessing socialexchange relationships; SERS ! Social Exchange Relationship Scale.! p # .05.
cative behavior definition. For example, the .12 value in the toprow of Table 2 shows that the perceived support scale had a .12higher level of definitional correspondence when paired with thesocial exchange relationship definition rather than the benefitdefinition. Because the 95% confidence interval for that differenceincluded .00, however, it is marked as not statistically significant.In contrast, the perceived support scale has a .52 higher level ofdefinitional correspondence when paired with the social exchangerelationship definition rather than the reciprocative behavior def-inition; this difference was statistically significant. Cohen’s dvalues are included in the table in parentheses next to each meandifference as a measure of effect size.The scale-level results are presented graphically in Figures 2, 3,
and 4, which again provide results for the complete sample, thesupervisor target condition, and the organization target condition.A content-valid pattern of results is evidenced in Table 2 andFigures 2–4 when a given scale’s mean definitional correspon-dence is statistically significantly higher for the social exchangerelationship definition than for both the benefit and reciprocativebehavior definitions. Results that adhere to that pattern are boldedin Table 2 and outlined in black in Figures 2–4.Three scales adhered to the content-valid pattern in both the
overall sample and each of the two target conditions: ABT,Bernerth, and SERS. Four other scales adhered to the content-validpattern in the overall sample and with the supervisor target, just notthe organization target: exchange quality, WBV, and Shore. Af-fective commitment adhered to the content-valid pattern only forthe overall sample. Finally, perceived support, CBT, and PCFnever adhered to the content-valid pattern.1Focusing solely on the three scales that adhered to the content-
valid pattern in all tests, we further examined whether there weredifferences between ABT, Bernerth, and SERS in terms of theircorrespondence with just the social exchange relationship defini-tion itself (i.e., just the middle bars in Figures 2–4). Post hoccomparisons showed that Bernerth had significantly higher levelsof correspondence than ABT in the overall sample (Mdiff ! .28,p # .05, Cohen’s d ! .25) and the supervisor target condition(Mdiff ! .33, p # .05, Cohen’s d ! .29). Post hoc comparisonsshowed that SERS had significantly higher levels of correspon-dence than ABT in the overall sample (Mdiff ! .21, p # .05,Cohen’s d ! .20) and the organization target condition (Mdiff !.22, p # .05, Cohen’s d ! .22). Post hoc comparisons found nosignificant differences in the levels of correspondence betweenBernerth and SERS in the overall sample, the supervisor targetcondition, and the organization target condition.
Supplementary AnalysesAlthough our Hinkin and Tracey (1999) analyses allowed us to
draw clear distinctions among the social exchange relationshipoperationalizations, it remains an open question whether otherkinds of analyses would surface similar implications. We thereforesupplemented our results with a confirmatory factor analysis-basedinvestigation, with participants recruited with classified ads postedon the Internet. Participants received $5 for completing a one-timesurvey and had to be 18 or older, to work for at least 35 hours perweek, and to not be self-employed. A total of 890 participantsregistered, with 691 completing the full survey. The participantswere 43% male, had an average age of 35.9 years (SD ! 11.78),
and reported the following ethnicities: 58% Caucasian, 15% Af-rican American, 11% Hispanic, 9% Asian, and 7% Other. Theyhad an average tenure with their organization of 5.35 years (SD !5.67) and an average tenure with their supervisor of 3.21 years(SD ! 3.45).Participants completed the survey online and were randomly
assigned to either a supervisor target condition or an organizationtarget condition. The survey included all of the scales in theAppendix along with an operationalization of both benefits andreciprocative behavior. Justice (either supervisory or organiza-tional) was used to operationalize benefits and was indicated bythree overall fairness scales (Ambrose & Schminke, 2009; Choi,2008; Colquitt, Long, Rodell, & Halvorsen-Ganepola, 2013). Cit-izenship behavior (either OCBS or OCBO) was used to operation-alize reciprocative behavior and was indicated by a helping scale(Van Dyne & LePine, 1998), a voice scale (Van Dyne & LePine,1998), and a civic virtue scale (Podsakoff, MacKenzie, Moorman,& Fetter, 1990). The survey was administered in three blocks—justice, social exchange, and citizenship behavior—with the order-ing of the blocks randomized. Scale order was then randomizedwithin blocks, with items presented in their published orders.Our analysis focused generally on whether fit issues in our
measurement modeling would echo the results observed in ourHinkin and Tracey (1999) results. To examine this possibility, wecontrasted two different measurement models. Model 1 was abaseline model with three latent variables: (a) benefits, indicatedby the three overall fairness scales; (b) social exchange relation-ships, indicated by perceived support, exchange quality, affectivecommitment, WBV, ABT, CBT, PCF, Shore, Bernerth, and SERS;and (c) reciprocative behavior, indicated by helping, voice, andcivic virtue. Model 2 then took that baseline and used the item-level Hinkin and Tracey (1999) results in Appendix A to alter theloadings of some items. For example, the results for the first CBTitem showed that it had significantly higher definitional correspon-dence for the benefits definition than for the social exchangerelationship definition. That item therefore loaded only on thebenefits factor and not the social exchange relationships factor. Asanother example, the results for the first perceived support itemshowed that it had equivalent definitional correspondence for thebenefits definition and for the social exchange relationship defini-tion. That item was therefore allowed to load on both the benefitsfactor and the social exchange relationships factor.In contrasting Model 1 with Model 2, we collapsed across the
supervisor and organization targets to boost statistical power. Theresulting comparison is shown in Table 3. Model 2 fit the databetter than Model 1 using a chi-square difference test, a compar-
1 Although scholars have argued that intellectual ability is the mostcritical requirement for participants in content validity studies (Hinkin &Tracey, 1999; Schriesheim et al., 1993), it is possible that employedparticipants could see more nuance in social exchange concepts that couldalter their perceptions of definitional correspondence. To examine thisissue, we asked participants to indicate whether they were currently em-ployed and how many hours per week they worked. Ninety of the 234participants in our study were employed, working an average of 17.33 hrper week (SD ! 8.18 hr). Use of employment status as a moderator in ourrepeated-measures ANOVAs failed to yield significant product terms foreither an employed versus unemployed dummy variable or a continuoushours worked variable. Those results suggest that our content validityresults are robust to the employment status of the participant.
ison of confidence intervals for the root mean square error ofapproximation (RMSEA), the Akaike information criterion (AIC),and side-by-side contrasts of other fit indices. The superior fit ofModel 2 illustrates that the Hinkin and Tracey (1999) results surfaceditem-level distinctions that were supported by our factor analyses.Items that seemed to load (or cross-load) on other definitions in ourHinkin and Tracey (1999) analyses also seemed to load (or cross-load) on other latent variables in our measurement model.
DiscussionTheory and measurement tend to proceed in one of a few
different paths in industrial and organizational psychology. Some-times, a theory is introduced after empirical testing with measuresor manipulations has already occurred (e.g., Ajzen, 1991; Deci &Ryan, 1985; Locke & Latham, 1990; Tajfel & Turner, 1986).
Other times, the introduction of a theory is followed quickly byempirical testing by the theory’s authors or by others (Vroom,1964; Weiss & Cropanzano, 1996). Still other times, the theoryand its operationalizations are introduced more or less concur-rently (Adams, 1965; Bandura, 1977; Hackman & Oldham, 1976).None of these paths were followed with contemporary socialexchange theory. Organ (1988, 1990) shone a spotlight on con-cepts that were articulated a quarter century earlier, in an effort tounderstand citizenship behavior. Other scholars—especially justicescholars—then gravitated to those concepts to understand linkageswith what would become that literature’s modal criterion (e.g., Kon-ovsky & Pugh, 1994; Masterson et al., 2000; Moorman et al., 1998).Given those scholars’ focus on what were still relatively new inde-pendent and dependent variables, it is perhaps not surprising that lessattention was paid to the content validity of their chosen mediators.
Figure 2. Definitional correspondence levels for overall sample (n ! 234). Scales that exhibit the content-validpattern to a statistically significant degree are outlined in black. Ben ! benefit definition; SER ! socialexchange relationship definition; Rec! reciprocative behavior definition; WBV! willingness to be vulnerable;ABT ! affect-based trust; CBT ! cognition-based trust; PCF ! psychological contract fulfillment; Shore !Shore et al. (2006) scale assessing social exchange relationships; Bernerth ! Bernerth et al. (2007) scaleassessing social exchange relationships; SERS ! Social Exchange Relationship Scale.
What stands out most from our results is that the most oft-utilized indicator of social exchange relationships—perceived sup-port (and, especially, POS)—was not shown to be content valid byHinkin and Tracey’s (1999) procedure. More than any other mea-sure in our study, perceived support samples not just from thesocial exchange relationships conceptual domain but also from theconceptual domain of benefits—of actions by an exchange partnerthat create a desire to reciprocate. In Blau’s (1964) formulation,such benefits include assistance and appreciation; with perceivedsupport they include helping, recognition, and consideration(Eisenberger et al., 1986). In hindsight, this result is understand-able because Eisenberger et al. (1986) never intended perceivedsupport to be an indicator of social exchange relationships butrather of the kind of treatment that would engender affectivecommitment. It was other scholars who cast it as a mediator in the
benefit ¡ social exchange relationship ¡ reciprocative behaviorcausal string (e.g., Masterson et al., 2000; Moorman et al., 1998;Tekleab et al., 2005), even though such studies wind up resemblinga benefit ¡ benefit ¡ reciprocative behavior string.Although the second most frequent operationalization of social
exchange relationships—exchange quality (and, especially,LMX)—fit the content-valid pattern in some tests, it did not drawfrom the content domain as strongly as other options. Concernsabout the content of exchange quality echo Bernerth et al.’s (2007)criticisms, when they introduced their scale, that the seminalexchange quality measures were created before the literature be-came tightly associated with social exchange theory. That realitymakes the content of some of the items (i.e., knowing where onestands, having a partner understand one’s needs) somewhat dis-connected from the beliefs and sentiments that Blau (1964) used to
Figure 3. Definitional correspondence levels for supervisor sample (n ! 112). Scales that exhibit thecontent-valid pattern to a statistically significant degree are outlined in black. Ben ! benefit definition; SER !social exchange relationship definition; Rec ! reciprocative behavior definition; PSS ! perceived supervisorsupport; LMX ! leader–member exchange; WBV ! willingness to be vulnerable; ABT ! affect-based trust;CBT ! cognition-based trust; PCF ! psychological contract fulfillment; Shore ! Shore et al. (2006) scaleassessing social exchange relationships; Bernerth ! Bernerth et al. (2007) scale assessing social exchangerelationships; SERS ! Social Exchange Relationship Scale.
describe social exchange relationships, even as other items evokebenefits and reciprocative behavior content.In contrast, our results showed that one of the more infrequently
utilized mediators, McAllister’s (1995) ABT, was among the mostcontent-valid indicators of social exchange relationships. Thatsuperior validity may be a testament to the fact that McAllister(1995) was influenced by Clark and Mills’s (1979) discussion ofcommunal relationships, a conceptualization that has been likenedto Blau’s (1964) social exchange relationship (Cropanzano &Mitchell, 2005). Regardless, the ABT items seem to explicitlyevoke a relationship while also tapping the sentiments at play inBlau’s (1964) theorizing—all without inadvertently sampling ben-efits or reciprocative behavior content. In their review of theliterature, Cropanzano and Mitchell (2005) noted that the findingsregarding trust as an indicator of social exchange mediation were
“promising” but also “sparse” (p. 886). Our results suggest thatsuch “sparseness” should be addressed using ABT, given that CBTdid not fit the content-valid pattern and WBV did so to a lesserextent (and only in the overall and supervisor target testing).Other than ABT, the only scales that fit the content-valid pat-
terns across all tests were two of the direct measures: Bernerth etal. (2007) and the SERS. Bernerth et al. (2007) fit the pattern byevoking relationships, having hints of the sentiments described byBlau (1964), and keeping any mentions of benefits or reciprocationsomewhat vague and undefined. It is somewhat surprising thatBernerth et al.’s (2007) scale remains so underutilized, given itsperformance in our tests. Other than being due to the recency ofits introduction, we suspect, that infrequency is a function of itscosmetic similarity to exchange quality scales and Shore et al.’s(2006) scale. Its content looks similar to those more established—
Figure 4. Definitional correspondence levels for organization sample (n ! 122). Scales that exhibit thecontent-valid pattern to a statistically significant degree are outlined in black. Ben ! benefit definition; SER !social exchange relationship definition; Rec ! reciprocative behavior definition; POS ! perceived organiza-tional support; OMX ! organization–member exchange; WBV ! willingness to be vulnerable; ABT !affect-based trust; CBT! cognition-based trust; PCF! psychological contract fulfillment; Shore! Shore et al.(2006) scale assessing social exchange relationships; Bernerth ! Bernerth et al. (2007) scale assessing socialexchange relationships; SERS ! Social Exchange Relationship Scale.
and therefore less controversial—choices, even though the itemsdo a better job of capturing relationships without evoking benefitsand reciprocative behavior. The SERS, in contrast, fit the patternby focusing its instructions solely on relationships and using themutual obligation, trust, commitment, and significance sentimentsdescribed by Blau (1964) as its items. Those results, together withits strong reliability and shorter length, would seem to make it aworthy option for future exchange-based research. One other op-tion, suggested by a reviewer, would be utilizing an amalgam ofitems that fit the content-valid pattern, taken from separate scales.That sort of amalgam is shown in Appendix B, though we shouldcaution that it may be controversial to use published items outsideof their published scale structures.Of course, a natural question becomes, Which of the three
content-valid options is the best choice for scholars in their re-search? Although future empirical testing will certainly inform theanswer to that question, it seems that ABT would be especiallyappropriate when scholars are seeking to contribute to both thetrust literature and the social exchange literature. If not, bothBernerth et al.’s (2007) scale and the SERS had significantlystronger content validity in the overall sample. Of those two,Bernerth et al.’s (2007) scale would be more appropriate whenscholars are seeking to contribute to both the LMX/OMX literatureand the social exchange literature. Given its shorter length, theSERS would likely be useful in cases where survey space is at apremium or participant fatigue is a concern (see Donnellan, Os-wald, Baird, & Lucas, 2006, for a discussion of such issues).
Limitations and ImplicationsThis study has some limitations that should be noted. For
example, we confined our examination of the social exchangeindicators to supervisor and organization targets. Although thoseare the two most common targets in tests of social exchangetheorizing, scholars have begun to examine other targets. Forexample, scholars have referenced both exchange quality andaffective commitment to coworkers (e.g., Lavelle et al., 2009; Tse& Dasborough, 2008), typically in an effort to predict citizenshiptargeted at coworkers, often termed OCBI (Williams & Anderson,1991). It remains unclear how our findings would generalize tosuch additional targets. Second, our study utilized undergraduatebusiness students, who may have less familiarity with social ex-change contexts. It should be noted, however, that content knowl-
edge and familiarity are not a requirement of Hinkin and Tracey’s(1999) technique. Moreover, employment status was not found tobe a significant moderator of our results.Despite these limitations, our study offers important implica-
tions for one of the most oft-evoked theories in industrial andorganizational psychology. In reflecting on the importance offalsifiability in theory testing, Platt (1964) recounted a quote froma noted biologist: “A theory which cannot be mortally endangeredcannot be alive” (p. 349). It seems difficult to endanger socialexchange theory—in terms of empirically refuting its core propo-sitions (Bacharach, 1989)—when those propositions can be sup-ported using any of 10 different mediators. That issue only getsexacerbated when some of those mediators are infused with thevery benefit and reciprocative behavior concepts that they aresupposed to be linking. We hope that our critical examination ofthe operationalizations of social exchange relationships will bringmore focus and clarity to the testing of the theory’s propositionsmoving forward.
ReferencesAdams, J. S. (1965). Inequity in social exchange. In L. Berkowitz (Ed.),Advances in experimental social psychology (Vol. 2, pp. 267–299). NewYork, NY: Academic Press.
Ajzen, I. (1991). The theory of planned behavior. Organizational Behaviorand Human Decision Processes, 50, 179–211. doi:10.1016/0749-5978(91)90020-T
Allen, D. G., Shore, L. M., & Griffeth, R. W. (2003). The role of perceivedorganizational support and supportive human resource practices in theturnover process. Journal of Management, 29, 99–118. doi:10.1177/014920630302900107
Allen, N. J., & Meyer, J. P. (1990). The measurement and antecedents ofaffective, continuance and normative commitment. Journal of Occupa-tional Psychology, 63, 1–18. doi:10.1111/j.2044-8325.1990.tb00506.x
Ambrose, M. L., & Schminke, M. (2009). The role of overall justicejudgments in organizational justice research: A test of mediation. Jour-nal of Applied Psychology, 94, 491–500. doi:10.1037/a0013203
Aryee, S., Budhwar, P. S., & Chen, Z. X. (2002). Trust as a mediator of therelationship between organizational justice and outcomes: Test of asocial exchange model. Journal of Organizational Behavior, 23, 267–285. doi:10.1002/job.138
Bacharach, S. B. (1989). Organizational theories: Some criteria for eval-uation. Academy of Management Review, 14, 496–515.
Table 3Goodness of Fit Comparisons
Goodness of fit statisticModel 1
(baseline model)Model 2
(Hinkin & Tracey, 1999, adjusted)
Chi-square 11,840.41! 10,245.71!
• Degrees of freedom 2207 2156• Chi-square difference 1,594.7!
Root-mean-square error of approximation (RMSEA) .105 .085• 90% confidence interval [.103, .106] [.083, .086]Akaike information criterion (AIC) 19,128.25 13,220.67Comparative fit index (CFI) .97 .98Root-mean-square residual (RMR) .094 .069Goodness of fit index (GFI) .55 .65
Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioralchange. Psychological Review, 84, 191–215. doi:10.1037/0033-295X.84.2.191
Bernerth, J. B., Armenakis, A. A., Feild, H. S., Giles, W. F., & Walker,H. J. (2007). Leader–member social exchange (LMSX): Developmentand validation of a scale. Journal of Organizational Behavior, 28,979–1003. doi:10.1002/job.443
Berry, D. T. R., Wetter, M. W., Baer, R. A., Larsen, L., Clark, C., &Monroe, K. (1992). MMPI-2 random responding indices: Validationusing a self-report methodology. Psychological Assessment, 4, 340–345.doi:10.1037/1040-3590.4.3.340
Blau, P. M. (1964). Exchange and power in social life. New York, NY:Wiley.
Choi, J. (2008). Event justice perceptions and employees’ reactions: Per-ceptions of social entity justice as a moderator. Journal of AppliedPsychology, 93, 513–528. doi:10.1037/0021-9010.93.3.513
Clark, M. E., Gironda, R. J., & Young, R. W. (2003). Detection of backrandom responding: Effectiveness of MMPI-2 and Personality Assess-ment Inventory validity indices. Psychological Assessment, 15, 223–234. doi:10.1037/1040-3590.15.2.223
Clark, M. S., & Mills, J. (1979). Interpersonal attraction in exchange andcommunal relationships. Journal of Personality and Social Psychology,37, 12–24. doi:10.1037/0022-3514.37.1.12
Colquitt, J. A. (2001). On the dimensionality of organizational justice: Aconstruct validation of a measure. Journal of Applied Psychology, 86,386–400. doi:10.1037/0021-9010.86.3.386
Colquitt, J. A., LePine, J. A., Piccolo, R. F., Zapata, C. P., & Rich, B. L.(2012). Explaining the justice–performance relationship: Trust as ex-change deepener or trust as uncertainty reducer? Journal of AppliedPsychology, 97, 1–15. doi:10.1037/a0025208
Colquitt, J. A., Long, D. M., Rodell, J. B., & Halvorsen-Ganepola, M. D.K. (2013). Adding the “in” to justice: A qualitative and quantitativeinvestigation of the differential effects of justice and injustice. Manu-script submitted for publication.
Coyle-Shapiro, J. A. M., & Conway, N. (2004). The employment relation-ship through the lens of social exchange. In J. A. M. Coyle-Shapiro,L. M. Shore, M. S. Taylor, & L. Tetrick (Eds.), The employmentrelationship: Examining psychological and contextual perspectives (pp.5–28). Oxford, United Kingdom: Oxford University Press.
Cropanzano, R., & Byrne, Z. S. (2000). Workplace justice and the dilemmaof organizational citizenship. In M. Van Vugt, T. Tyler, & A. Biel(Eds.), Collective problems in modern society: Dilemmas and solutions(pp. 142–161). London, United Kingdom: Routledge.
Cropanzano, R., & Mitchell, M. S. (2005). Social exchange theory: Aninterdisciplinary review. Journal of Management, 31, 874–900. doi:10.1177/0149206305279602
Cropanzano, R., Prehar, C. A., & Chen, P. Y. (2002). Using socialexchange theory to distinguish procedural from interactional justice.Group & Organization Management, 27, 324–351. doi:10.1177/1059601102027003002
Cropanzano, R., & Rupp, D. E. (2008). Social exchange theory andorganizational justice: Job performance, citizenship behaviors, multiplefoci, and a historical integration of two literatures. In S. W. Gilliland,D. P. Skarlicki, & D. D. Steiner (Eds.), Research in social issues inmanagement: Justice, morality, and social responsibility (pp. 63–99).Greenwich CT: Information Age.
Cropanzano, R., Rupp, D. E., & Byrne, Z. S. (2003). The relationship ofemotional exhaustion to work attitudes, job performance, and organiza-tional citizenship behaviors. Journal of Applied Psychology, 88, 160–169. doi:10.1037/0021-9010.88.1.160
Cropanzano, R., Rupp, D. E., Mohler, C. J., & Schminke, M. (2001). Threeroads to organizational justice. In G. R. Ferris (Ed.), Research inpersonnel and human resource management (Vol. 19, pp. 1–113). NewYork, NY: Elsevier Science.
Deci, E. L., & Ryan, R. M. (1985). Intrinsic motivation and self-determination in human behavior. New York, NY: Plenum Press.
Donnellan, M. B., Oswald, F. L., Baird, B. M., & Lucas, R. E. (2006). TheMini-IPIP scales: Tiny-yet-effective measures of the Big Five factors ofpersonality. Psychological Assessment, 18, 192–203. doi:10.1037/1040-3590.18.2.192
Eby, L. T., Allen, T. D., Hoffman, B. J., Baranik, L. E., Sauer, J. B.,Baldwin, S., . . . Evans, S. C. (2013). An interdisciplinary meta-analysisof the potential antecedents, correlates, and consequences of protégéperceptions of mentoring. Psychological Bulletin, 139, 441–476. doi:10.1037/a0029279
Eisenberger, R., Armeli, S., Rexwinkel, B., Lynch, P. D., & Rhoades, L.(2001). Reciprocation of perceived organizational support. Journal ofApplied Psychology, 86, 42–51. doi:10.1037/0021-9010.86.1.42
Eisenberger, R., Huntington, R., Hutchison, S., & Sowa, D. (1986). Per-ceived organizational support. Journal of Applied Psychology, 71, 500–507. doi:10.1037/0021-9010.71.3.500
Eisenberger, R., Karagonlar, G., Stinglhamber, F., Neves, P., Becker, T. E.,Gonzalez-Morales, M. G., & Steiger-Mueller, M. (2010). Leader–member exchange and affective organizational commitment: The con-tribution of supervisor’s organizational embodiment. Journal of AppliedPsychology, 95, 1085–1103. doi:10.1037/a0020858
Eisenberger, R., Stinglhamber, F., Vandenberghe, C., Sucharski, I. L., &Rhoades, L. (2002). Perceived supervisor support: Contributions toperceived organizational support and employee retention. Journal ofApplied Psychology, 87, 565–573. doi:10.1037/0021-9010.87.3.565
El Akremi, A., Vandenberghe, C., & Camerman, J. (2010). The role ofjustice and social exchange relationships in workplace deviance: Test ofa mediated model. Human Relations, 63, 1687–1717. doi:10.1177/0018726710364163
Foa, E., & Foa, U. (1980). Resource theory: Interpersonal behavior asexchange. In K. J. Gergen, M. S. Greenberg, & R. H. Willis (Eds.),Social exchange: Advances in theory and research (pp. 77–94). NewYork, NY: Plenum Press.
Folger, R., & Cropanzano, R. (2001). Fairness theory: Justice as account-ability. In J. Greenberg & R. Cropanzano (Eds.), Advances in organi-zational justice (pp. 1–55). Stanford, CA: Stanford University Press.
Gerstner, C. R., & Day, D. V. (1997). Meta-analytic review of leader–member exchange theory: Correlates and construct issues. Journal ofApplied Psychology, 82, 827–844. doi:10.1037/0021-9010.82.6.827
Graen, G. B., & Uhl-Bien, M. (1995). Relationship-based approach toleadership: Development of leader–member exchange (LMX) theory ofleadership over 25 years: Applying a multi-level, multi-domain perspec-tive. Leadership Quarterly, 6, 219–247. doi:10.1016/1048-9843(95)90036-5
Greenberg, J., & Scott, K. S. (1996). Why do workers bite the hands thatfeed them? Employee theft as a social exchange process. In B. M. Staw& L. L. Cummings (Eds.), Research in organizational behavior (Vol.18, pp. 111–156). Greenwich, CT: JAI Press.
Hackman, J. R., & Oldham, G. R. (1976). Motivation through the design ofwork: Test of a theory. Organizational Behavior and Human Perfor-mance, 16, 250–279. doi:10.1016/0030-5073(76)90016-7
Hinkin, T., & Tracey, J. (1999). An analysis of variance approach tocontent validation. Organizational Research Methods, 2, 175–186. doi:10.1177/109442819922004
Hom, P. W., Tsui, A. S., Wu, J. B., Lee, T. W., Zhang, A. Y., Fu, P. P., &Li, L. (2009). Explaining employment relationships with social ex-change and job embeddedness. Journal of Applied Psychology, 94,277–297. doi:10.1037/a0013453
Huang, X., Iun, J., Liu, A., & Gong, Y. (2010). Does participative lead-ership enhance work performance by inducing empowerment or trust?The differential effects on managerial and non-managerial subordinates.Journal of Organizational Behavior, 31, 122–143. doi:10.1002/job.636
James, L. R., Demaree, R. G., & Wolf, G. (1984). Estimating within-groupinterrater reliability with and without response bias. Journal of AppliedPsychology, 69, 85–98. doi:10.1037/0021-9010.69.1.85
Karriker, J. H., & Williams, M. L. (2009). Organizational justice andorganizational citizenship behavior: A mediated multifoci model. Jour-nal of Management, 35, 112–135. doi:10.1177/0149206307309265
Keppel, G., & Wickens, T. D. (2004). Design and analysis: A researcher’shandbook (4th ed.). Upper Saddle River, NJ: Pearson.
Kerlinger, F. N., & Lee, H. B. (2000). Foundations of behavioral research.Forth Worth, TX: Harcourt.
Kickul, J. R., Neuman, G., Parker, C., & Finkl, J. (2001). Settling the score:The role of organizational justice in the relationship between psycho-logical contract breach and anticitizenship behavior. Employee Respon-sibilities and Rights Journal, 13, 77–93. doi:10.1023/A:1014586225406
Konovsky, M. A., & Pugh, S. D. (1994). Citizenship behavior and socialexchange. Academy of Management Journal, 37, 656–669. doi:10.2307/256704
Kozlowski, S. W. J. (2012). The Oxford handbook of organizationalpsychology. New York, NY: Oxford University Press.
Lavelle, J. J., Brockner, J., Konovsky, M. A., Price, K. H., Henley, A. B.,Taneja, A., & Vinekar, V. (2009). Commitment, procedural fairness, andorganizational citizenship behavior: A multifoci analysis. Journal ofOrganizational Behavior, 30, 337–357. doi:10.1002/job.518
Lee, K., Scandura, T., Kim, Y., Joshi, K., & Lee, J. Y. (2012). Examiningleader–member exchange as a moderator of the relationship betweenemotional intelligence and creativity of software developers. Engineer-ing Management Research, 1, 15–28. doi:10.5539/emr.v1n1p15
Lester, S. W., Kickul, J. R., & Bergmann, T. J. (2007). Managing employeeperceptions of the psychological contract over time: The role of em-ployer social accounts and contract fulfillment. Journal of Organiza-tional Behavior, 28, 191–208. doi:10.1002/job.410
Liao, H., Joshi, A., & Chuang, A. (2004). Sticking out like a sore thumb:Employee dissimilarity and deviance at work. Personnel Psychology, 57,969–1000. doi:10.1111/j.1744-6570.2004.00012.x
Liden, R. C., Wayne, S. J., & Stillwell, D. (1993). A longitudinal study onthe early development of leader–member exchanges. Journal of AppliedPsychology, 78, 662–674. doi:10.1037/0021-9010.78.4.662
Locke, E. A., & Latham, G. P. (1990). A theory of goal setting and taskperformance. Englewood Cliffs, NJ: Prentice-Hall.
Malatesta, R. M., & Byrne, Z. S. (1997, April). The impact of formal andinteractional procedures on organizational outcomes. Paper presented atthe conference of the Society for Industrial and Organizational Psychol-ogy, St. Louis, MO.
Masterson, S. S., Lewis, K., Goldman, B. M., & Taylor, M. S. (2000).Integrating justice and social exchange: The differing effects of fairprocedures and treatment on work relationships. Academy of Manage-ment Journal, 43, 738–748. doi:10.2307/1556364
Mayer, R. C., & Davis, J. H. (1999). The effect of the performanceappraisal system on trust for management: A field quasi-experiment.Journal of Applied Psychology, 84, 123–136. doi:10.1037/0021-9010.84.1.123
Mayer, R. C., Davis, J. H., & Schoorman, F. D. (1995). An integrativemodel of organizational trust. Academy of Management Review, 20,709–734.
Mayer, R. C., & Gavin, M. B. (2005). Trust in management and perfor-mance: Who minds the shop while the employees watch the boss?Academy of Management Journal, 48, 874–888. doi:10.5465/AMJ.2005.18803928
McAllister, D. J. (1995). Affect- and cognition-based trust as foundationsfor interpersonal cooperation in organizations. Academy of ManagementJournal, 38, 24–59. doi:10.2307/256727
McAllister, D. J., Lewicki, R. J., & Chaturvedi, S. (2006). Trust in
developing relationships: From theory to measurement. In K. M. Weaver(Ed.), Academy of Management proceedings (pp. G1–G6). doi:10.5465/AMBPP.2006.22897235
Meade, A. W., & Craig, S. B. (2012). Identifying careless responses insurvey data. Psychological Methods, 17, 437–455. doi:10.1037/a0028085
Meyer, J. P., Allen, N. J., & Smith, C. A. (1993). Commitment to orga-nizations and occupations: Extension and test of a three-componentconceptualization. Journal of Applied Psychology, 78, 538–551. doi:10.1037/0021-9010.78.4.538
Montes, S. D., & Irving, P. G. (2008). Disentangling the effects of prom-ised and delivered inducements: Relational and transactional contractelements and the mediating role of trust. Journal of Applied Psychology,93, 1367–1381. doi:10.1037/a0012851
Moorman, R. H., Blakely, G. L., & Niehoff, B. P. (1998). Does organiza-tional support mediate the relationship between procedural justice andorganizational citizenship behavior? Academy of Management Journal,41, 351–357. doi:10.2307/256913
Mowday, R. T., Porter, L. W., & Steers, R. M. (1982). Employee–organization linkages: The psychology of commitment, absenteeism, andturnover. New York, NY: Academic Press.
Organ, D. W. (1988). Organizational citizenship behavior: The goodsoldier syndrome. Lexington, MA: Lexington Books.
Organ, D. W. (1990). The motivational basis of organizational citizenshipbehavior. In L. L. Cummings & B. M. Staw (Eds.), Research in orga-nizational behavior (Vol. 12, pp. 43–72). Greenwich, CT: JAI Press.
Platt, J. R. (1964, October 16). Strong inference. Science, 146, 347–353.doi:10.1126/science.146.3642.347
Podsakoff, P. M., MacKenzie, S. B., Lee, J., & Podsakoff, N. P. (2003).Common method bias in behavioral research: A critical review of theliterature and recommended remedies. Journal of Applied Psychology,88, 879–903. doi:10.1037/0021-9010.88.5.879
Podsakoff, P. M., MacKenzie, S. B., Moorman, R. H., & Fetter, R. (1990).Transformational leader behaviors and their effects on followers’ trust inleader, satisfaction, and organizational citizenship behaviors. LeadershipQuarterly, 1, 107–142. doi:10.1016/1048-9843(90)90009-7
Robinson, S. L., Kraatz, M. S., & Rousseau, D. M. (1994). Changingobligations and the psychological contract: A longitudinal study. Acad-emy of Management Journal, 37, 137–152. doi:10.2307/256773
Robinson, S. L., & Morrison, E. W. (1995). Psychological contracts andOCB: The effect of unfulfilled obligations on civic virtue behavior.Journal of Organizational Behavior, 16, 289–298. doi:10.1002/job.4030160309
Robinson, S. L., & Morrison, E. W. (2000). The development of psycho-logical contract breach and violation: A longitudinal study. Journal ofOrganizational Behavior, 21, 525–546. doi:10.1002/1099-1379(200008)21:5#525::AID-JOB40$3.0.CO;2-T
Robinson, S. L., & Rousseau, D. M. (1994). Violating the psychologicalcontract: Not the exception but the norm. Journal of OrganizationalBehavior, 15, 245–259. doi:10.1002/job.4030150306
Rosen, C. C., Chang, C.-H., Johnson, R. E., & Levy, P. E. (2009).Perceptions of the organizational context and psychological contractbreach: Assessing competing perspectives. Organizational Behavior andHuman Decision Processes, 108, 202–217. doi:10.1016/j.obhdp.2008.07.003
Rousseau, D. M. (1990). New hire perceptions of their own and theiremployer’s obligations: A study of psychological contracts. Journal ofOrganizational Behavior, 11, 389–400. doi:10.1002/job.4030110506
Rupp, D. E., & Cropanzano, R. (2002). The mediating effects of socialexchange relationships in predicting workplace outcomes from multifociorganizational justice. Organizational Behavior and Human DecisionProcesses, 89, 925–946. doi:10.1016/S0749-5978(02)00036-5
Scandura, T. A., & Graen, G. B. (1984). Moderating effects of initialleader–member exchange status on the effects of a leadership interven-
tion. Journal of Applied Psychology, 69, 428–436. doi:10.1037/0021-9010.69.3.428
Schmitt, N., & Stults, D. M. (1985). Factors defined by negatively keyeditems: The result of careless respondents. Applied Psychological Mea-surement, 9, 367–373. doi:10.1177/014662168500900405
Schoorman, F. D., Mayer, R. C., & Davis, J. H. (2007). An integrativemodel of organizational trust: Past, present, and future. Academy ofManagement Review, 32, 344–354. doi:10.5465/AMR.2007.24348410
Schriesheim, C. A., Powers, K. J., Scandura, T. A., Gardiner, C. C., &Lankau, M. J. (1993). Improving construct measurement in managementresearch: Comments and a quantitative approach for assessing the the-oretical adequacy of paper-and-pencil and survey-type instruments.Journal of Management, 19, 385–417. doi:10.1016/0149-2063(93)90058-U
Schwab, D. P. (1980). Construct validity in organizational behavior. InL. L. Cummings & B. M. Staw (Eds.), Research in organizationalbehavior (Vol. 2, pp. 3–43). Greenwich, CT: JAI Press.
Shore, L. M., Tetrick, L. E., Lynch, P., & Barksdale, K. (2006). Social andeconomic exchange: Construct development and validation. Journal ofApplied Social Psychology, 36, 837–867. doi:10.1111/j.0021-9029.2006.00046.x
Sidak, Z. (1967). Rectangular confidence regions for the means of multi-variate normal distributions. Journal of the American Statistical Asso-ciation, 62, 626–633.
Song, L. J., Tsui, A. S., & Law, K. S. (2009). Unpacking employeeresponses to organizational exchange mechanisms: The role of socialand economic exchange perceptions. Journal of Management, 35, 56–93. doi:10.1177/0149206308321544
Tajfel, H., & Turner, J. C. (1986). The social identity theory of intergroupbehavior. In S. Worchel & W. G. Austin (Eds.), Psychology of inter-group relations (pp. 7–24). Chicago, IL: Nelson Hall.
Tekleab, A. G., Takeuchi, R., & Taylor, M. S. (2005). Extending the chainof relationships among organizational justice, social exchange, andemployee reactions: The role of contract violations. Academy of Man-agement Journal, 48, 146–157. doi:10.5465/AMJ.2005.15993162
Tekleab, A. G., & Taylor, M. S. (2003). Aren’t there two parties in anemployment relationship? Antecedents and consequences of organization–
employee agreement on contract obligations and violations. Journal of Orga-nizational Behavior, 24, 585–608. doi:10.1002/job.204
Tse, H. H. M., & Dasborough, M. T. (2008). A study of exchange andemotions in team member relationships. Group & Organization Man-agement, 33, 194–215. doi:10.1177/1059601106293779
Van Dyne, L., & LePine, J. A. (1998). Helping and voice extra-rolebehaviors: Evidence of construct and predictive validity. Academy ofManagement Journal, 41, 108–119. doi:10.2307/256902
Vroom, V. H. (1964). Work and motivation. New York, NY: Wiley.Walumbwa, F. O., Cropanzano, R., & Hartnell, C. A. (2009). Organiza-tional justice, voluntary learning behavior, and job performance: A testof the mediating effects of identification and leader–member exchange.Journal of Organizational Behavior, 30, 1103–1126. doi:10.1002/job.611
Wayne, S. J., Shore, L. M., Bommer, W. H., & Tetrick, L. E. (2002). Therole of fair treatment and rewards in perceptions of organizationalsupport and leader–member exchange. Journal of Applied Psychology,87, 590–598. doi:10.1037/0021-9010.87.3.590
Wayne, S. J., Shore, L. M., & Liden, R. C. (1997). Perceived organiza-tional support and leader–member exchange: A social exchange per-spective. Academy of Management Journal, 40, 82–111. doi:10.2307/257021
Weiss, H. M., & Cropanzano, R. (1996). Affective events theory: Atheoretical discussion of the structure, causes and consequences ofaffective experiences at work. In B. M. Staw & L. L. Cummings (Eds.),Research in organizational behavior (Vol. 18, pp. 1–74). Greenwich,CT: JAI Press.
Williams, L. J., & Anderson, S. E. (1991). Job satisfaction and organiza-tional commitment as predictors of organizational citizenship behaviors.Journal of Management, 17, 601–617. doi:10.1177/014920639101700305
Yang, J., Mossholder, K. W., & Peng, T. K. (2009). Supervisory proceduraljustice effects: The mediating roles of cognitive and affective trust.Leadership Quarterly, 20, 143–154. doi:10.1016/j.leaqua.2009.01.009
Zhao, H., Wayne, S. J., Glibkowski, B. C., & Bravo, J. (2007). The impactof psychological contract breach on work-related outcomes: A meta-analysis. Personnel Psychology, 60, 647–680. doi:10.1111/j.1744-6570.2007.00087.x
Appendix A
Scale Items for Variables Used as Indicators of Social Exchange Relationships
Scales and scale itemsSERmean rWG
OverallF
Follow-up meandifferences
SER vs.Ben
SER vs.Rec
Perceived support (Eisenberger et al., 2001)(Benefit % ! .83; SER % ! .75; reciprocative behavior
% ! .87)• My supervisor/organization takes pride in myaccomplishments. 5.13 .62 10.21! .06 (.04) .41! (.29)
% ! .92)• I usually know where I stand with mysupervisor/organization. 4.67 .51 11.20! .33! (.22) .39! (.27)
• My supervisor/organization understands my problemsand needs well enough. 4.73 .62 20.13! .38! (.28) .56! (.40)
• My supervisor/organization recognizes my potentialsome but not enough. (R) 3.71 .54 2.78 .18 (.13) .22 (.17)
• Regardless of how much power mysupervisor/organization has built, he or she/it would beinclined to use his or her/its power to help me solveproblems at work. 4.72 .56 16.23! .17 (.12) .62! (.43)
• I can count on my supervisor/organization to “bail meout” at his or her/its expense when I really need it. 4.64 .47 18.92! .26 (.17) .69! (.44)
• I have enough confidence in mysupervisor/organization to defend and justify his orher/its decisions when he or she/management is notpresent to do so. 4.94 .49 2.28 .21 (.14) .21 (.14)
• My working relationship with mysupervisor/organization is effective. 5.14 .58 11.41! .38! (.27) .42! (.30)
Affective commitment (Meyer et al., 1993)(Benefit % ! .82; SER % ! .81; reciprocative behavior
% ! .79)• I would be very happy to spend the rest of my careerwith this supervisor/organization. 4.70 .20 3.91! .21 (.12) .33! (.19)
• I really feel as if this supervisor’s/organization’sproblems are my own. 4.71 .43 5.14! .33! (.21) .29 (.18)
• My supervisor/This organization has a great deal ofpersonal meaning for me. 4.74 .36 1.56 .17 (.10) .18 (.11)
• I do not feel like “part of the family” with mysupervisor/organization. (R) 3.31 .24 3.98! .24 (.14) .33! (.20)
• I do not feel “emotionally attached” to my supervisor/organization. (R) 3.42 .26 3.91! .24 (.14) .29! (.17)
• I do not feel a strong sense of “belonging” with mysupervisor/to my organization. (R) 3.21 .20 2.44 .21 (.12) .23 (.14)
Willingness to be vulnerable (Schoorman et al., 2007)(Benefit % ! .80; SER % ! .75; reciprocative behavior
% ! .85)• My supervisor/organization keeps my interests in mindwhen making decisions. 4.87 .59 20.80! .10 (.08) .63! (.44)
• I would be willing to let my supervisor/organizationhave significant influence over my future in thiscompany. 4.62 .43 3.98! .16 (.11) .30! (.19)
• If my supervisor/management asked why a problemoccurred, I would speak freely even if I were partly toblame. 4.74 .48 5.48! .32! (.22) .27! (.18)
• I feel comfortable being creative because mysupervisor/organization understands that sometimescreative solutions do not work. 4.28 .38 2.17 .12 (.07) .23 (.14)
• It is important for me to have a good way to keep aneye on my supervisor/management. (R) 3.43 .48 1.49 .16 (.11) .12 (.08)
• Increasing my vulnerability to criticism by mysupervisor/organization would be a mistake. (R) 3.33 .43 5.01! .24 (.17) .31! (.21)
• If I had my way, I wouldn’t let mysupervisor/organization have any influence overdecisions that are important to me. (R) 3.23 .37 3.86! .28! (.18) .24 (.14)
% ! .91)• My supervisor/organization and I have a sharingrelationship. My supervisor/management and I canfreely share our ideas, feelings, and hopes. 5.28 .60 15.11! .37! (.28) .55! (.40)
• I can talk freely to my supervisor/management aboutdifficulties I am having at work and know that mysupervisor/management will want to listen. 5.12 .56 8.29! .30! (.21) .40! (.28)
• My supervisor/organization and I would both feel asense of loss if I could no longer work there. 4.75 .41 4.85! .31! (.20) .30! (.20)
• If I shared my problems with mysupervisor/organization, I know that my supervisor/management would respond constructively andcaringly. 5.11 .59 12.50! .30! (.22) .51! (.36)
• I would have to say that my supervisor/organizationand I have both made considerable emotionalinvestments in our working relationship. 5.08 .49 10.34! .36! (.25) .46! (.31)
% ! .94)• My supervisor/organization approaches his or her job/its business with professionalism and dedication. 4.41 .11 17.21! ".25! (.14) .32! (.17)
• Given my supervisor’s/organization’s track record, Isee no reason to doubt his or her/its competence andpreparation. 4.37 .22 10.28! ".07 (.04) .33! (.18)
• I can rely on my supervisor/organization not to makemy job more difficult by carelessness. 4.16 .29 15.85! ".25 (.15) .40! (.23)
• Most people, even those who aren’t very familiar withmy supervisor/organization, trust and respect him orher/it [as a fellow employee]. 4.36 .20 8.05! ".13 (.08) .26! (.14)
• Other work associates of mine [who must interactwith my supervisor] consider him or her/myorganization to be trustworthy. 4.54 .24 14.80! ".03 (.02) .45! (.25)
• If people knew more about my supervisor/thisorganization and his or her/its background, they wouldbe more concerned and monitor my supervisor’s/itsperformance more closely. (R) 2.78 .47 .45 .06 (.04) .10 (.07)
• Almost all of the promises made by my supervisor/organization have been kept so far. 5.09 .63 41.38! .08 (.06) .81! (.57)
• I feel that my supervisor/organization has comethrough in fulfilling the promises made to me when Iwas hired. 5.15 .62 37.26! ".07 (.06) .73! (.51)
• So far my supervisor/organization has done anexcellent job of fulfilling his or her/its promises tome. 5.31 .58 44.13! ".03 (.02) .85! (.55)
• I have not received everything promised to me inexchange for my contributions. (R) 3.59 .30 7.96! .20 (.12) .47! (.30)
• My supervisor/organization has broken many of his orher/its promises to me even though I’ve upheld myside of the deal. (R) 3.26 .11 8.60! .21 (.12) .49! (.29)
Shore (Shore et al., 2006)(Benefit % ! .86; SER % ! .84; reciprocative behavior
% ! .86)• My supervisor/organization has made a significantinvestment in me. 5.15 .55 25.65! .05 (.04) .66! (.44)
• The things I do on the job today will benefit mystanding with this supervisor/organization in the longrun. 5.24 .63 5.83! .33! (.26) .20 (.15)
• There is a lot of give and take in my relationship withmy supervisor/organization. 5.12 .55 5.50! .21 (.16) .30! (.22)
• I worry that all my efforts on behalf of my supervisor/organization will never be rewarded. (R) 3.71 .27 4.51! .33! (.20) .11 (.07)
• I don’t mind working hard today—I know I willeventually be rewarded by my supervisor/organization. 5.12 .60 3.93! .12 (.09) .29! (.21)
• My relationship with my supervisor/organization isbased on mutual trust. 5.27 .61 15.78! .33! (.24) .53! (.39)
• I try to look out for the best interest of my supervisor/organization because I can rely on mysupervisor/organization to take care of me. 5.33 .61 3.95! .10 (.08) .26! (.20)
• Even though I may not always receive the recognitionfrom my supervisor/organization I deserve, I know myefforts will be rewarded in the future. 4.87 .60 1.86 .20 (.15) .10 (.07)
Bernerth (Bernerth et al., 2007)(Benefit % ! .94; SER % ! .94; reciprocative behavior
% ! .93)• My supervisor/organization and I have a two-wayexchange relationship. 5.67 .64 10.70! .37! (.30) .28! (.23)
• I do not have to specify the exact conditions to knowmy supervisor/organization will return a favor. 5.36 .59 19.70! .47! (.35) .66! (.49)
• If I do something for my supervisor/organization, mysupervisor/organization will eventually repay me. 5.26 .55 8.89! .29! (.22) .40! (.29)
Note. n ! 234. Coefficient alpha is reported for each definitional condition. (R) denotes a negatively worded item. SER!social exchange relationship definition; Ben ! benefit definition; Rec ! reciprocative behavior definition.! p # .05.
Scale Items for Amalgamated Measure of Social Exchange Relationships
Scale itemSERmean rWG
OverallF
Follow-up meandifferences
SER vs.Ben
SER vs.Rec
(Benefit % ! .92; SER % ! .88; reciprocativebehavior % ! .92)• My supervisor/organization and I have a two-way exchange relationship. (Bernerth) 5.67 .64 10.70! .37! (.30) .28! (.23)
• My relationship with my supervisor/organizationis composed of comparable exchanges of givingand taking. (Bernerth) 5.41 .58 7.52! .29! (.23) .34! (.26)
• My efforts are reciprocated by my supervisor/organization. (Bernerth) 5.37 .65 5.79! .22! (.18) .32! (.24)
• I do not have to specify the exact conditions toknow my supervisor/organization will return afavor. (Bernerth) 5.36 .59 19.70! .47! (.35) .66! (.49)
• My supervisor/organization and I have a sharingrelationship. My supervisor/management and Ican freely share our ideas, feelings, and hopes.(ABT) 5.28 .60 15.11! .37! (.28) .55! (.40)
• My relationship with my supervisor/organizationis based on mutual trust. (Shore) 5.27 .61 15.78! .33! (.24) .53! (.39)
• If I do something for mysupervisor/organization, mysupervisor/organization will eventually repayme. (Bernerth) 5.26 .55 8.89! .29! (.22) .40! (.29)
• My working relationship with my supervisor/organization is effective. (Exchange Quality) 5.14 .58 11.41! .38! (.27) .42! (.30)
• I can talk freely to my supervisor/managementabout difficulties I am having at work and knowthat my supervisor/management will want tolisten. (ABT) 5.12 .56 8.29! .30! (.21) .40! (.28)
• If I shared my problems with my supervisor/organization, I know that mysupervisor/management would respondconstructively and caringly. (ABT) 5.11 .59 12.50! .30! (.22) .51! (.36)
• I would have to say that mysupervisor/organization and I have both madeconsiderable emotional investments in ourworking relationship. (ABT) 5.08 .49 10.34! .36! (.25) .46! (.31)
• I have a balance of inputs and outputs with mysupervisor/organization. (Bernerth) 5.06 .59 6.82! .29! (.22) .33! (.25)
Note. n ! 234. The scale from which the items were taken is noted in parentheses after each item. SER !social exchange relationship definition; Ben ! benefit definition; Rec ! reciprocative behavior definition;Bernerth ! Bernerth et al. (2007) scale assessing social exchange relationships; ABT ! affect-based trust;Shore ! Shore et al. (2006) scale assessing social exchange relationships.! p # .05.
Received September 7, 2012Revision received February 17, 2014
Correction to Colquitt et al. (2014)In the article “Scale Indicators of Social Exchange Relationships: A Comparison of RelativeContent Validity,” by Jason A. Colquitt, Michael D. Baer, David M. Long, and Marie D. K.Halvorsen-Ganepola (Journal of Applied Psychology, Advance online publication. April 7, 2014.doi:10.1037/a0036374), the first sentence in the caption for Figure 4 should have been: “Defini-tional correspondence levels for organization sample (n !122).” All versions of this article havebeen corrected.
