WHAT WILL THE BOSS THINK? THE IMPRESSION...

PERSONNEL PSYCHOLOGY2015, 68, 463–498

WHAT WILL THE BOSS THINK? THE IMPRESSIONMANAGEMENT IMPLICATIONS OF SUPPORTIVERELATIONSHIPS WITH STAR AND PROJECT PEERS

DAVID M. LONGCollege of William and Mary

MICHAEL D. BAERUniversity of Georgia

JASON A. COLQUITTUniversity of Georgia

RYAN OUTLAWUniversity of Georgia

RASHPAL K. DHENSA-KAHLONAston University

Although impression management scholars have identified a number oftactics for influencing supervisor evaluations, most of those tactics rep-resent supervisor-targeted behaviors. This study examines the degreeto which employees form supportive relationships with peers for im-pression management purposes. In so doing, we explore this intriguingquestion: Will employees gain more from forming supportive relation-ships with “stars” (i.e., top performers who are “on the fast track” inthe organization) or “projects” (i.e., “works in progress” who need helpand refinement to perform well)? We examined this question in 2 fieldstudies. Study 1 included 4 sources and 2 time periods; Study 2 included2 sources and 3 time periods. The results showed that supportive rela-tionships with both stars and projects seemed to represent impressionmanagement opportunities, insofar as they predicted supervisor positiveaffect and perceptions of employee promotability. Impression manage-ment motives only predicted supportive relationships with stars, how-ever, not projects. Relationships with projects were driven by prosocialmotives not concerns about managing images. We discuss the practicaland theoretical implications of our results for the managing of impres-sions and peer relationships.

How one is perceived by others is a prevalent concern for most individ-uals (Gardner & Martinko, 1988; Leary, 1995; Leary & Kowalski, 1990).This contention is particularly true in organizational life given that the

Correspondence and requests for reprints should be addressed to David M. Long, MasonSchool of Business, College of William and Mary, 3074 Miller Hall, Williamsburg, VA23187; [email protected]⃝ 2014 Wiley Periodicals, Inc. doi: 10.1111/peps.12091

463

464 PERSONNEL PSYCHOLOGY

“others” include supervisors who control job assignments, rewards, andpromotions. It is therefore not surprising that employees attempt to controlsupervisor perceptions. Research on impression management is devotedto understanding the tactics employees use to shape those perceptions (forreviews, see Bolino, Kacmar, Turnley, & Gilstrap, 2008; DuBrin, 2011;Rosenfeld, Giacalone, & Riordan, 1995). For the most part, scholars havefocused on tactics that occur within the employee–supervisor relation-ship, such as giving compliments, pointing out their own achievements,or ensuring that bosses see how hard they are working (Bolino et al., 2008;Ferris et al., 2002; Higgins, Judge, & Ferris, 2003; Rosenfeld et al., 1995).

Although those tactics are certainly an important piece of the im-pression management puzzle, supervisor perceptions may also be shapedby actions and events occurring in other work relationships. For exam-ple, research suggests that being perceived as having a prominent friendwithin the organization can increase supervisor perceptions of employeeperformance (Kilduff & Krackhardt, 1994). As another example, studieshave shown that being connected to well-regarded others can have pos-itive effects on how employees are viewed (Andrews & Kacmar, 2001;Cialdini, 1989; Richardson & Cialdini, 1981). Such results hint at the im-pression management implications of employee relationships with peers.Of course, it must be noted that employees have relationships with anumber of different peers—peers who vary in how valued they are andhow well-regarded they seem to be. Impression management scholarshave argued that “what we do is often less important than whom we doit with” (Cialdini & Richardson, 1980, p. 194). If so, to what degree doemployees view relationships with different kinds of peers as impressionmanagement opportunities?

Our research builds and tests theory about the impression managementimplications of employee relationships with peers. Although relationshipsare multifaceted, we build our theory around the concept of supportiverelationships, which we define as relationships where support and as-sistance are provided to others. Unlike other types of relationships thatmay develop through more formal organizational channels (e.g., team-based interdependencies), supportive relationships are discretionary onthe part of employees. They therefore represent a potential impressionmanagement tactic (Bolino, 1999). Moreover, these types of relationshipsmay be especially relevant to employees’ images, as scholars have notedthat providing support to a coworker can bring praise and recognitionto employees (Allen, Poteet, & Burroughs, 1997; Allen & Rush, 1998;Baumeister, 1982; Bolino, 1999; Bolino, Varela, Bande, & Turnley, 2006).

In particular, our theorizing argues that supportive relationshipswith peers can positively influence two different supervisor evaluations.One of those evaluations is positive affect, which reflects supervisors’

DAVID M. LONG ET AL. 465

enthusiasm, excitement, and positive sentiments toward employees(Frijda, 1994; Watson, Clark, & Tellegen, 1988). The other is perceivedpromotability, which captures supervisor evaluations of employee ad-vancement potential (Harris, Kacmar, & Carlson, 2006; Thacker & Wayne,1995). Ultimately, these two evaluations are the bottom-line gauge of theeffectiveness of employee impression management (Bozeman & Kacmar,1997; Ferris & Judge, 1991; Gardner & Martinko, 1988). If employeesare well liked and viewed as promotable, then it is likely that they havemanaged impressions effectively.

Our research examines supportive relationships with two distinct typesof peers. One type is a star, defined here as a top performer who seemsto be “on the fast track” in the organization and is expected to make aclear contribution to the firm. Star employees make significant contribu-tions to the performance of the organization yet are often stretched totheir capacity in order to capitalize on their talents (Oldroyd & Morris,2012). The other type is a project, defined here as someone who seems tobe a “work in progress” who needs help and refinement to become a topperformer. Our use of that term is consistent with talent evaluations in theworld of sports, where projects are described as individuals with “highceilings, only their elevators are still on the ground floor or lower levels”(Wasserman, 2014). Our use is also consistent with one of the broadermeanings of the project term—an undertaking requiring concerted effort(The American Heritage Dictionary). Relative to stars, supportive rela-tionships with projects represent a more uncertain risk–reward equation.Supporting projects may reveal a true “diamond in the rough” but mayalso represent an inefficient use of limited time and effort.

Our theorizing highlights a potential irony in the use of supportiverelationships with peers as an impression management tool. On the onehand, employees should be more likely to view the support of stars as aneffective impression management tactic, given the benefits of connectingwith well-regarded others (Allen, 2004; Cialdini, 1989). On the otherhand, there are reasons to expect the support of projects to be just aspredictive of positive affect on the part of the supervisor. To the degreethat this is so, supporting projects may be as valuable a road to perceivedpromotability as supporting stars. Indeed, finding similar outcomes forsupporting stars and projects would have important implications for peersand supervisors. If employees who support projects avoid “missing out”on impression management opportunities, the peers who need support themost will be more likely to receive it. This more extensive network ofsupport should benefit supervisors, who may need help in developing andsupporting projects. We examined these issues in two field studies withdifferent recruitment approaches, different methods for identifying starsand projects, and different operationalizations of supportive relationships.


Impression Management Motives and Supportive Relationships

Most employees are aware of the impact that their supervisors’ percep-tions have on outcomes such as job assignments, rewards, and promotions(Feldman & Klich, 1991; Gardner & Martinko, 1988; Rosenfeld et al.,1995). It is that awareness that leads to an interest in making a favorableimpression on supervisors (Bolino, 1999; Bolino et al., 2008; Kim, VanDyne, Kamdar, & Johnson, 2013; Leary & Kowalski, 1990; Rioux &Penner, 2001; Roberts, 2005). This interest is represented by employeeimpression management motives—desires to adapt behavior to projecta good image (Leary, 1995; Yun, Takeuchi, & Liu, 2007). Leary andKowalski (1990) suggested that such desires represent the first of a two-step impression management process. The second step is for employeesto construct their desired image through impression-forming behaviors.

Although those impression-forming behaviors can center on tacticslike self-promotion or ingratiation (Bolino et al., 2008; Ferris et al., 2002;Higgins et al., 2003; Rosenfeld et al., 1995), scholars have argued thatemployees believe offering encouragement, feedback, and support to theircoworkers can be beneficial to their image (Allen et al., 1997; Newby &Heide, 1992; Noe, Greenberger, & Wang, 2002) and that such behaviorscan result in rewards from supervisors (e.g., Allen, Lentz, & Day, 2006).Indeed, employees could be drawn to supportive relationships with peersas an impression management approach because it is more subtle thantactics like self-promotion or ingratiation—both of which can risk neg-ative labels such as “egotist” or “bootlicker” (DuBrin, 2011; Gardner &Martinko, 1998; Ham & Vonk, 2011; Jones, 1990; Jones & Pittman, 1982;Leary, 1995). In our first study, we operationalize supportive relationshipsas psychosocial support—the provision of respect, encouragement, con-firmation, and feedback (Kram, 1985; Kram & Isabella, 1985).

A key question in our study is whether employee impression man-agement motives will predict both supportive relationships with stars andsupportive relationships with projects, or whether they will instead onlypredict the former. According to impression management theory, em-ployees should assume that supervisors will think highly of them for es-tablishing relationships with well-regarded others (Cialdini, 1989, 2001;Leary, 1995; Schlenker, 1980). In support of such theorizing, researchhas shown that employees are more inclined to help others who are in-fluential in the organization (Bowler & Brass, 2006). Other work sug-gests that perceptions of popularity and skill increase for those who areconnected to high-status others (Dijkstra, Cillessen, Lindenberg, & Veen-stra, 2010; Kilduff & Krackhardt, 1994). Indeed, impression managementtheory suggests that building supportive relationships with stars can pro-vide employees with the opportunity to “bask in the reflected glory” of


well-regarded others (Cialdini & de Nicholas, 1989; Cialdini & Richard-son, 1980; Crandall, Silvia, N’Gbala, Tsang, & Dawson, 2007).

In contrast, the instrumentality of supportive relationships withprojects may be less clear to employees. Although projects have the po-tential to perform well in the organization, they do not possess the samelevel of talent and skill as their star counterparts. As such, employees mayview projects as not offering the same opportunity to bask in the reflectedglory of a well-regarded other (Cialdini & Richardson, 1980). Indeed,impression management scholars have suggested that employees mightdownplay or avoid relationships with certain peers to avoid having su-pervisors associate them with less-regarded others (Andrews & Kacmar,2001; Cialdini, 1989; Kulik, Bainbridge, & Cregan, 2008; Leary, 1995).Moreover, it is not clear that supportive relationships with projects would“pay off” down the line, in terms of high-quality reciprocative exchangesof resources between colleagues. Research suggests that such exchangescan have an important impact on job attitudes (Sherony & Green, 2002).

Hypothesis 1: Impression management motives are positively relatedto supportive relationships with stars.

Supportive Relationships and Supervisor Evaluations

Regardless of whether (and why) employees may engage in support-ive relationships with stars and projects, the next question becomes howthose relationships can shape supervisor evaluations of them. In gen-eral, impression management behaviors have been shown to affect howsupervisors feel toward—and think about—the employees who performthem (Bolino et al., 2008; Gardner & Martinko, 1988; Rosenfeld et al.,1995). Do those findings extend to a context where the behaviors arepeer directed, rather than supervisor directed? Our study examines thatquestion by examining two different evaluations: positive affect and per-ceived promotability.

Positive Affect

Perhaps the central outcome of employee impression managementefforts is positive affect on the part of supervisors (Bolino et al., 2006;Cialdini, 1989; Crandall et al., 2007; Ferris & Judge, 1991; Ferris, Judge,Rowland, & Fitzgibbons, 1994; Kacmar & Carlson, 1999; Wayne & Ferris,1990; Wayne & Liden, 1995). Impression management theory argues thatpositive affect can be fostered when supervisors perceive that they benefitfrom the tactics employed (Bolino et al., 2008; Cialdini, 1989; Ferris& Judge, 1991). That same logic is consistent with appraisal models of


affect. For example, Lazarus’s (2008) model argues that positive affectoccurs when events are perceived to be beneficial to one’s goals and one’sidentity. The question becomes, then, whether positive affect should resultfrom both support of stars and support of projects.

To examine this question, it is useful to consider the kinds of goals andidentity elements that are considered in theory and research on affect. No-vacek and Lazarus (1990) identified power/achievement and stress avoid-ance as two broad, widely held goals. Lazarus (1991) later highlightedseveral aspects of the power/achievement goal: wanting to obtain awardsand recognition, aspiring to be well-off financially, and wanting to com-pete successfully. From the perspective of supervisors, power/achievementgoals can be achieved by producing, developing, and supporting top talent(Hunt & Michael, 1983; Lewis & Heckman, 2006; Noe, 2002). Employ-ees who are known to nurture supportive relationships with stars thereforefacilitate supervisors’ goal achievement by expanding the base of sup-port for that top talent (Hunt & Michael, 1983). With respect to identity,Lazarus (1991) argued that positive affect would be fostered when eventsimprove self- or social esteem, or when they reaffirm moral values, ideals,and meanings. We propose that the reflected glory obtained by employeeswho support stars is itself relevant to the self- and social-esteem of super-visors. Supervisors may feel more pride and enthusiasm when interactingwith employees who support stars precisely because that reflected gloryimbues those employees with more status.

Hypothesis 2: Supportive relationships with stars is positively relatedto positive affect on the part of supervisors.

Employees’ supportive relationships with projects may also be seen asgoal relevant by supervisors, with the operative goal being stress avoid-ance rather than power/achievement. Lazarus (1991) highlighted severalaspects of the stress avoidance goal: desiring an easy life, wanting toavoid blame or criticism, trying to avoid conflict, and wanting to avoidstress. Supervisors are responsible for the performance of the workgroup(Floyd & Wooldridge, 1994; Noe, 2002), and projects—whose futureperformance is uncertain—could be a source of stress for supervisors.Witnessing subordinates’ struggles can act as a stressor in its own right.However, projects’ struggles can also engender role conflict as supervisorsbalance assisting projects with investing time and attention in more proxi-mal drivers of unit performance. Employees could ease some of that strainby providing support to projects, thereby facilitating the achievement ofsupervisors’ stress avoidance goals. With respect to identity, employeeswho support projects may appeal to supervisors’ moral values and ideals,with positive affect being triggered by seeing someone support a per-son in need. Indeed, research on impression management suggests that


supervisors tend to like employees who help those in need (Bolino et al.,2006; Eastman, 1994).

Hypothesis 3: Supportive relationships with projects is positively re-lated to positive affect on the part of supervisors.

Perceived Promotability

From an instrumental perspective, an increase in positive affect on thepart of supervisors is beneficial to employees if it impacts career-relatedoutcomes. One such outcome is supervisor evaluations of employees’perceived promotability. Although research has previously linked the pro-vision of support to the number of employee promotions received (e.g.,Allen, 2007; Allen et al., 2006; Bozionelos, 2004; Hunt & Michael, 1983),the intervening mechanism explaining why support might result in per-ceived promotability has not been explored. We theorize that a supervisor’spositive affect may serve as that mediator.

One of the more venerable findings in the impression managementliterature is that supervisors who hold positive affect toward employeesmore positively evaluate their potential (e.g., Ferris & Judge, 1991; Ferriset al., 1994; Wayne & Ferris, 1990; Wayne & Liden, 1995). For exam-ple, research has demonstrated that supervisors who like employees as aresult of their impression efforts ultimately rate those employees higheron promotability (Shaughnessy, Treadway, Breland, Williams, & Brouer,2011). Scholars have suggested that positive affect can influence promot-ability by altering how supervisors process information about subordinates(Cardy & Dobbins, 1986; DeNisi & Williams, 1988; Ferris et al., 1994;Robbins & DeNisi, 1994). Specifically, positive affect impacts the storageand recall of information (Isen, Shalker, Clark, & Karp, 1978; Thacker &Wayne, 1995), leading supervisors to focus more on positively valencedinformation and less on negatively valenced information. Positive affectmay also mask employee deficiencies, creating a halo bias that extends toperceived promotability (Balzer & Sulsky, 1992; Ferris & Judge, 1991;Tsui & Barry, 1986).

Hypothesis 4: Supportive relationships with stars has a positive indi-rect effect on perceived promotability through positiveaffect on the part of supervisors.

Hypothesis 5: Supportive relationships with projects has a positiveindirect effect on perceived promotability through pos-itive affect on the part of supervisors.


Study 1: Method

Sample and Procedure

We recruited participants via classified advertisements posted on theInternet in 18 metropolitan areas across the United States. The advertise-ments indicated that the study was open to full-time employees who hada supervisor. In order to be eligible, employees must have worked with atleast one peer who “will clearly be a top performer in the organization”and one peer who “needs some help and refinement to be a top performerin the organization.” We took three steps to ensure that these eligibilityrequirements were met when employees selected their two peers. First,we emphasized the eligibility requirements in the advertisements. Second,employees were directed to a web page detailing the study requirementsand listing definitional descriptions of stars and projects, though thosespecific labels were never utilized. Finally, at the beginning of the survey,employees completed a “selection check” in which they indicated the ex-tent to which the peers they had identified conformed to the definitionaldescriptions of stars and projects provided on the web page.

Given that our conceptualizations of stars and projects had not beenpreviously established in the literature, we created and validated scalesthat could be used in our selection checks. We developed these scales us-ing measure creation and validation procedures recommended by Hinkinand Tracey (1999; also see Hinkin, 1998). Adhering to the conceptualdefinitions of stars and projects outlined earlier, we created 10 items de-signed to reflect both peers. We then recruited an independent sample ofundergraduate students (N = 139) from a large southeastern university toquantitatively assess the extent to which these items matched the concep-tual definitions of stars and projects. Undergraduates are an appropriatesample for Hinkin and Tracey’s (1999) validation method because partici-pants only require sufficient intellectual ability to rate the convergence be-tween items and definitions (Hinkin & Tracey, 1999; Schriesheim, Powers,Scandura, Gardiner, & Lankau, 1993). Indeed, recent studies have beenpublished in top psychology and management journals using the techniquewith undergraduates (Colquitt, Baer, Long, & Halvorsen-Ganepola, 2014;Rodell, 2013). The students were randomly assigned to one of two onlinesurvey conditions. Seventy-five students received a survey that providedthe star definition, followed by the 10 star items and the 10 project items.Sixty-four students received a survey that provided the project definition,followed by the 10 project items and the 10 star items. To minimize itemcontext effects, the order in which the items were presented to the studentswas randomized.


Students were asked to rate the extent to which each item provideda good match to the definition provided using this scale: 1 = extremelybad match to the definition to 7 = extremely good match to the definition.The items with the highest means on the intended definition—and thebest ability to draw distinctions with the unintended definition—wereretained for use in the selection check measures. Those criteria netted fiveitems for the star selection check and five items for the project selectioncheck. The five star items were: “Has ‘what it takes’ to go far in myorganization,” “Will clearly make a contribution to my organization,”“Will climb the ladder quickly,” “Is on the ‘fast track’,” and “Seemslike a ‘winner’.” The mean definitional correspondence for these fiveitems was 5.52 out of 7. The five project items were: “Could succeedwith some assistance,” “Could be described as a ‘work in progress’,”“Has underdeveloped capabilities but shows promise,” “Shows glimpsesof talent,” and “May not be the best yet but has demonstrated an ability tolearn.” The mean definitional correspondence for these five items was 5.20out of 7. Those levels of definitional correspondence compare favorably topast uses of this technique and illustrate adequate content validity for ourselection checks (Colquitt et al., 2014; Hinkin & Tracey, 1999; Rodell,2013).

A total of 1,077 employees clicked on our advertisements. Once theydid so, they were shown the study description and eligibility requirementsand were asked to provide the contact information for their supervisor, astar peer, and a project peer. A total of 432 employees self-selected into thestudy by providing that contact information. At that point, they began theTime 1 survey, which included the selection check items described above.Employees were asked, regarding their chosen star, “In my opinion, [peername] is someone who . . . ,” with the star selection check items using thefollowing scale: 1 = not at all to 5 = to a very large extent. Employeesfollowed the same procedure to rate their chosen project on the projectselection check. The survey also included the measure of impression man-agement motives for establishing relationships with the stars and projects.The survey was completed by 391 employees for a completion rate of88%, with employees receiving $5 for their participation. In all cases,payments were sent as cash through regular mail directly to the relevantparticipants. The employees were 42% male and, on average, were 31.3years old (SD = 9.23), had worked for their supervisor for 3.6 years (SD =3.64), and had been with their organization for 4.6 years (SD = 4.41). Theethnicity of the employees was 48% Caucasian, 18% African American,16% Hispanic, 11% Asian American, and 2% Native American, with theremainder indicating other ethnicities. The employees came from at least12 different industries, including manufacturing, healthcare, government,education, transportation, and retail.


Upon receiving a completed survey from the employees, we sent thestar and project peers an email inviting them to participate in the study. Im-portantly, that invitation came from us rather than allowing the employeesthemselves to contact the two peers. We felt this structure would lend morelegitimacy to the study while also allowing us to verify the independentidentities of the stars and projects. The peer surveys included the measureof supportive relationships. One-hundred and seventy-seven stars agreedto participate in the study, resulting in a response rate of 45%. The meanfor those 177 peers on the star selection check completed by employeeswas 4.43 out of 5, providing strong evidence that the employees pickedpeers who matched the star definition. Stars were 48% male and had anaverage age of 33.7 years (SD = 10.17). The ethnicity of the stars was50% Caucasian, 16% African American, 14% Hispanic, and 13% AsianAmerican, with the remainder indicating other ethnicities. On average,stars had been with their organization for 3.0 years (SD = 3.23). One-hundred and sixty-six projects agreed to participate in the study, resultingin a response rate of 42%. The mean for those 166 peers on the project se-lection check was 4.09 out of 5, again providing evidence that employeespicked peers who matched the project definition. The projects were 55%male and had an average age of 33.7 years (SD = 10.03). The ethnicity ofthe projects was 50% Caucasian, 17% African American, 16% Hispanic,and 10% Asian American, with the remainder indicating other ethnicities.On average, projects had been with their organization for 2.7 years (SD =2.59). The star and project peers were paid $5 for participating.

Six weeks after the completion of the employee survey, supervisorswere invited to participate in the survey. As with the star and projectsurveys, this communication came directly from us to increase legitimacyand verify the identity of the supervisors. The supervisor survey includedmeasures of positive affect and perceived promotability of the employee.Of the 391 supervisors who received the invitation to participate, 133responded, for a response rate of 34%. Supervisors were paid $5 fortheir participation. Supervisors had an average age of 41.1 years (SD =10.03) and were 52% male. On average, supervisors had worked for theirorganization for 8.1 years (SD = 5.91) and had 17 subordinates (SD =24.99). Supervisors were 61% Caucasian, 15% African American, 12%Hispanic, and 8% Asian American, with the remainder indicating otherethnicities.

Complete data for the “quartets,” which includes the employee survey,a star survey, a project survey, and a supervisor survey, were available for125 employees. Attrition across times and sources was combatted withweekly reminder emails, including soliciting help from other members ofa given “quartet” when a given source was late in completing their portion.


Unless otherwise noted below, all surveys used a five-point scale where1 = strongly disagree to 5 = strongly agree.

Employee Measures

Impression management motives. Impression management motives forsupportive peer relationships were measured with the Yun et al. (2007)six-item self-enhancement motives scale. Employees indicated the extentto which they agreed that working with the peer provided the opportunitiesindicated by the items. The employee responded to both star- and project-referenced versions of the scale. The introduction to each measure wasmodified to read, “Working with [peer name] is an opportunity to . . . .”The items included “Change my behavior to create a good impressionto others,” “Create the impression that I am a ‘good’ person to others,”“Modify my behavior to present a good image to others,” “Be sensitive tothe impression that others have of me,” “Present myself to others as beingfriendlier and more polite,” and “Demonstrate how important it is to meto give a good impression to others.” The coefficient alpha was .88 for thestar-referenced scale and .91 for the project-referenced scale.

Peer Measures

Supportive relationships. Stars and projects both assessed the supportthey received from the employee using the 10 psychosocial support itemsfrom Dreher and Ash’s (1990) mentoring scale. All items began with theintroduction, “[Employee name] has . . . ” The items included “Encour-aged me to talk openly about anxiety and fears that detract from work,”“Conveyed feelings of respect for me as an individual,” “Shared personalexperiences as an alternative perspective to my problems,” “Conveyed em-pathy for the concerns and feelings I discussed with him/her,” “Discussedmy questions or concerns regarding feelings of competence, commitmentto advancement, relationships with peers or supervisors, or work/familyconflicts,” “Shared history of his/her career with me,” “Served as a rolemodel,” “Encouraged me to prepare for advancement,” “Displayed atti-tudes and values similar to my own,” and “Encouraged me to try new waysof behaving on the job.” The coefficient alpha for supportive relationshipswas .94 for both the star and project peers.

Supervisor Measures

Positive affect. Positive affect toward the employee was measured us-ing the 10 positive affect adjectives from the positive and negative affect


schedule–expanded form (PANAS-X; Watson & Clark, 1994). Supervi-sors were asked to “Indicate to what extent you typically feel this waywhen thinking about or interacting with [employee name].” A five-pointscale (1 = not at all to 5 = to a very large extent) was used. The itemsincluded enthusiastic, excited, inspired, proud, strong, determined, in-terested, active, alert, and attentive. The coefficient alpha for this scalewas .92.

Perceived promotability. Perceptions of the employee’s overall pro-motability were measured with a seven-item scale from Harris et al.(2006). The items included “Is someone I would select as a successorto my position,” “Is someone I believe has high potential,” “Is the type ofindividual our company seeks to promote,” “Is someone that will have asuccessful career,” “Is someone I would approach if I needed advice,” “Issomeone that seems to ‘fit in’ well around here,” and “Is someone whoseopinions have an impact on my decisions.” This scale had a coefficientalpha of .91.

Control Variables

We collected employee ratings of the Big Five personalitydimensions—for use as potential control or moderating variables—usingDonnellan, Oswald, Baird, and Lucas’s (2006) four-item mini-IPIP (In-ternational Personality Item Pool) scales. We examined the moderatingeffects of conscientiousness (α = .67) and openness to experience (α =.75) to explore the possibility that the responders to our advertisementswere unusually motivated or unusually curious. We examined emotionalstability (α = .85) as a control given that it predicts “backing up be-havior” (Porter et al., 2003)—a construct somewhat similar to supportiverelationships with projects. During data collection, we supplemented theemotional stability scale with four items from the IPIP due to concerns thatthe four-item measure might exhibit low reliability. In the end, includingthese personality dimensions did not alter the results of our hypothesistesting. Our analyses therefore proceeded without them, consistent withrecent recommendations (Carlson & Wu, 2012). Demographic controls—including tenure with the organization, tenure with the supervisor, race,gender, and age—also did not alter our hypothesis testing, so they werenot included in our analyses.

Confirmatory Factor Analysis

A confirmatory factor analysis with impression management mo-tives for star, impression management motives for project, supportiverelationship with star, supportive relationship with project, positive affect,


TABLE 1Study 1: Descriptive Statistics, Reliabilities, and Correlations

Variable Mean SD 1 2 3 4 5 6

1. Impression management motivesfor star

3.99 .67 .88

2. Impression management motivesfor project

3.79 .79 .51∗ .91

3. Supportive relationship with star 3.89 .86 .25∗ .14 .944. Supportive relationship with project 3.74 .91 .27∗ .15 .53∗ .945. Positive affect 4.20 .63 .22∗ .11 .42∗ .44∗ .926. Perceived promotability 4.40 .52 .29∗ .26∗ .43∗ .38∗ .71∗ .91

Note. n = 125. Coefficient alphas are on the diagonal.∗p < .05.

and perceived promotability modeled as latent factors with item-levelindicators showed that our six-factor model provided a good fit to thedata: χ2 (1,065) = 1,981.75, p < .01; comparative fit index (CFI) = .94;incremental fit index (IFI) = .94; root mean square error of approxima-tion (RMSEA) = .076; standardized root mean square residual (SRMR) =.065. We compared our Study 1 measurement model to a five-factor modelin which the perceived promotability and positive affect items loaded ona single factor. That comparison revealed a χ2 difference of 102.11 (df =5, p < .001) in favor of our hypothesized measurement model. Thus, weretained our hypothesized model.

Results and Discussion

Descriptive Statistics

The descriptive statistics, coefficient alphas, and zero-order correla-tions for our variables are shown in Table 1.

Tests of Hypotheses

We tested our hypotheses using structural equation modeling in LIS-REL 8.80 (Joreskog & Sorbom, 1989). Latent variables were modeledwith item-level indicators. Direct effects from the supportive relation-ship variables to perceived promotability were also included becausethose paths are necessary for the calculation of indirect effects (MacK-innon, Lockwood, Hoffman, West, & Sheets, 2002). The resulting modelprovided an adequate fit to the data: χ2 (1,119) = 2,113.15, p < .01;


CFI = .94; IFI = .94; RMSEA = .076; SRMR = .14.1 Figure 1 shows thestandardized path coefficients from the LISREL output.

Hypothesis 1 predicted that impression management motives for starswould be positively related to supportive relationships with stars. Asshown in Figure 1, the relationship between impression management mo-tives for stars and supportive relationships with stars was significant (b =.27). Hypothesis 1 was therefore supported. Figure 1 also reveals a non-significant linkage between impression management motives for projectsand supportive relationships with projects.

Hypotheses 2 and 3 predicted that supportive relationships with starsand projects would be positively related to positive affect on the part ofsupervisors. As shown in Figure 1, the relationship between supportiverelationships with stars and positive affect was significant (b = .27),providing support for Hypothesis 2. Hypothesis 3 was also supported,as evidenced by the significant, positive relationship between supportiverelationships with projects and positive affect (b = .34).

Hypotheses 4 and 5 predicted that supportive relationships with starsand projects would have indirect effects on perceived promotabilitythrough positive affect. We tested these mediation hypotheses using aproduct of coefficients approach (MacKinnon et al., 2002). This ap-proach demonstrates mediation by showing that the product of thepredictor→mediator and mediator→outcome relationships is statisticallysignificant when direct effects are also modeled. Because the product oftwo path coefficients is rarely normally distributed, more accurate andpowerful tests of the significance of the mediated effect can be obtainedusing a distribution-of-the-product approach, which corrects for the non-normality of the product (MacKinnon, Fritz, Williams, & Lockwood,2007; Tofighi & MacKinnon, 2011). We conducted these distribution-of-the-product analyses using the RMediation package within R software(Tofighi & MacKinnon, 2011). The effect decomposition for these me-diated effects is shown in Table 2, which reveals that both supportiverelationships for stars (.18) and projects (.23) had significant indirect ef-fects on perceived promotability through positive affect, supporting bothpredictions. The magnitude of the indirect effects can be recovered usingthe tracing rule (Kline, 2011) with the Figure 2 coefficients (.27 * .67 =.18 for stars; .34 * .67 = .23 for projects).

Taken together, our results suggest that both supportive relationshipswith stars and supportive relationships with projects offer what might be

1We examined an alternative model where perceived promotability and positive affectwere modeled as dual terminal outcomes, as opposed to positive affect predicting perceivedpromotability. A comparison of the two models revealed a χ 2 difference of 54.20 (df = 1,p < .01) in favor of our hypothesized model. Thus, we retained our hypothesized structure.


Figu

re1:

Stru

ctur

alE

quat

ion

Mod

elin

gR

esul

tsfo

rSt

udy

1.

Not

e.n

=12

5.*p

<.0

5.


TABLE 2Study 1: Effect Decomposition Results for Effects of Supportive Relationships

With Stars and Projects on Perceived Promotability

Predictor

Supportive relationship Supportive relationshipwith star with project

Indirect effect throughPositive affect .18∗ .23∗

Direct effect .16∗ .00Total effect .34∗ .22∗

Note. n = 125.∗p < .05.

called “impression management opportunities.” That is, both supportiverelationships had their own indirect effects on perceived promotabilitythrough positive affect. Employees who engaged in both kinds of rela-tionships gained, from a perspective of perceived likeability and perceivedpotential. Importantly, those impression management opportunities onlyseemed salient to employees in the case of stars. Impression manage-ment motives did predict supportive relationships for stars but not forprojects. Finding similar outcomes for supportive relationships with starsand projects suggests that when it comes to “scoring points with theboss,” employees who could be supporting projects may be missing outon a valuable image-enhancing option.

Of course, Study 1 possessed some limitations. Although our recruit-ment approach resulted in a heterogeneous sample with some externalvalidity advantages (Shadish, Cook, & Campbell, 2002), it did not allowus to calculate a response rate because we could not track the numberof individuals who saw the advertisement without clicking on it. Our ap-proach also asked employees themselves to identify stars and projects,leaving open the question of whether our results would generalize to othernomination approaches. Similarly, we operationalized supportive relation-ships using psychosocial support (Dreher & Ash, 1990) when a numberof other constructs could serve as appropriate indicators. Finally, it maybe that our positive affect mediator is contaminated with other relevantsupervisor evaluations, such as perceived performance or perceived citi-zenship. Unpacking that possibility would make for a more compellingtest of the mediating role of positive affect. Study 2 addressed all ofthese limitations while exploring a question that was not answered inStudy 1: “Why would employees engage in supportive relationships withprojects?”


Study 2

If supportive relationships with projects do represent a missed “im-pression management opportunity” because of their effects on positiveaffect and perceived promotability, are some employees gaining thosebenefits for reasons other than impression management? Exploring thisissue requires a consideration of motives for behavior that do not centeron controlling supervisor evaluations. In a study of citizenship behavior,Rioux and Penner (2001) showed that discretionary behaviors could bedriven by prosocial behaviors, in addition to impression management be-haviors. For example, employees might maintain a positive work attitudeto be helpful to others and/or to impress their bosses.

Grant (2008) defined prosocial motives as a desire to protect and pro-mote the welfare of other people. His research illustrated that employeeswho want to make a prosocial difference engage in behaviors similar tothe supportive relationships examined in Study 1. For example, Grantand Mayer (2009) showed that prosocial motives and impression manage-ment motives both explained helping behaviors among peers. The authorsconcluded that “employees who are both good soldiers and good actorsare more likely to emerge as good citizens in promoting the status quo”(p. 900). In this study, we go one step further by suggesting that em-ployees who hold prosocial motives will consider the type of peer whenforming supportive relationships. Such employees should recognize thatthe welfare of projects is particularly in need of protection and promotion.Indeed, Grant and Gino (2010) showed that employees with prosocialmotives focused their attention on others when they felt their efforts couldmake a difference.

Hypothesis 6a: Impression management motives are positively re-lated to supportive relationships with stars.

Hypothesis 6b: Prosocial motives are positively related to supportiverelationships with projects.

We also hypothesize that our predictions in Hypotheses 2 through 5in Study 1 will replicate with the Study 2 sample. As in Study 1, findingsimilar outcomes for supportive relationships with stars and projects wouldlend support to our proposal that relationships with both types of peershave impression management benefits.

Method

We recruited participants through two different pools. One was a sam-ple of executive and professional MBA programs for a large southeastern


university. The other was a subject pool of a large southeastern universitythat included working undergraduates as well as other working adults notaffiliated with the university who had indicated a willingness to participatein university research projects. We contacted all 2,165 individuals in thesetwo pools to participate in this study. Those communications indicatedthat eligible participants must be working at least 20 hours per week,work in an environment with coworkers, and be willing to fill out twoonline surveys. Eligible individuals were asked to provide their contactinformation as well as the contact information for their supervisor. Ofthe 2,165 individuals who were contacted, 759 completed this employeeregistration survey, for a response rate of 35%.

We then contacted the 759 supervisors via email to invite them to par-ticipate in the study. Unlike the procedure in Study 1, the supervisors wereasked to read the definitional descriptions of stars and projects and to iden-tify one employee who fit the star description and one employee who fit theproject description. We also instructed the supervisors that these identifiedemployees should have regular interactions with the focal employee. Afterproviding the names of the star and project, the supervisors were askedto complete the selection check described in Study 1. Three-hundred andseventy-one supervisors completed this supervisor registration survey, fora response rate of 49%. The employees of these supervisors were thenemailed the Time 1 survey, which included measures of impression man-agement motives for stars, impression management motives for projects,prosocial motives for stars, prosocial motives for projects, and demo-graphic and control variables. Three-hundred and thirty-two employeescompleted the Time 1 survey, for a response rate of 89%. Four weeks later,these employees were emailed the Time 2 survey, which included mea-sures of supportive relationships with stars and supportive relationshipswith projects. The Time 2 survey was completed by 311 employees, for aresponse rate of 94%. Four weeks after the Time 2 survey, the supervisorsof those 311 employees were emailed the Time 3 survey, which includedmeasures of positive affect and perceived promotability, as well as de-mographic and control variables. Two-hundred and forty-two supervisorscompleted the Time 3 survey, for a response rate of 78%.

Complete data, which included a completed Time 1 and Time 2 sur-vey from the employee and a registration and Time 3 survey from thesupervisor, were available for 207 employee–supervisor dyads. At Time1, we included scales that assessed the extent to which the employee hadthe opportunity to observe the star and project peers that were selectedby the supervisor. Employees whose mean rating was below two on thefive-point scale for their opportunity to observe either the star or projectwere removed from the analyses. This criteria led to the removal of threedyads from our analyses, resulting in a final sample size of 204 dyads—an


effective response rate of 10%. Employees had an average age of 31.7 years(SD = 10.07) and were 50% male. They had worked for their organiza-tions for an average of 4.5 years (SD = 4.88) and for their supervisors foran average of 2.7 years (SD = 2.83). Employees identified their ethnicityas 61% Caucasian, 14% African American, 12% Asian/Pacific Islander,7% Hispanic, and 6% other. Seventy-one employees (35%) were them-selves in managerial roles, whereas 133 (65%) were not. Supervisors hadan average age of 39.6 years (SD = 10.99) and were 54% male. They hadworked for their organizations for an average of 7.9 years (SD = 6.77). Su-pervisors identified their ethnicity as 65% Caucasian, 12% Asian/PacificIslander, 11% African American, 6% Hispanic, and 6% other. The meanon the supervisors’ selection check for the stars was 4.40 out of 5. Themean on the selection check for the projects was 3.88 out of 5. All par-ticipants were paid $10 for participating in the study, with payments sentas cash through regular mail. Unless otherwise indicated, all scales belowused a five-point response scale from 1 = strongly disagree to 5 = stronglyagree.

Employee Measures

Impression management motives. Impression management motives forsupportive peer relationships were measured with Rioux and Penner’s(2001) 10-item impression management scale. As in Study 1, the employeeresponded to both star- and project-referenced versions of the scale. Theintroduction to each measure was modified to read, “Working with [peername] is an opportunity to . . . .” The items included “Look better thanmy other coworkers,” “Indicate that I want a raise,” “Impress my othercoworkers,” “Avoid looking bad in front of others,” “Look like I ambusy,” “Avoid appearing irresponsible,” “Avoid looking lazy,” “Stay outof trouble,” “Avoid a reprimand from my boss,” and “Show that rewardsare important to me.” The coefficient alpha was .95 for both the star-referenced scale and the project-referenced scale.

Prosocial motives. Employees’ prosocial motives for supportive peerrelationships were measured with Rioux and Penner’s (2001) 10-itemprosocial values scale. The items included “Help those in need,” “Showthat I am concerned about others’ feelings,” “Be helpful,” “Be courteousto others,” “Help my coworkers in any way I can,” “Interact with mycoworkers,” “Have fun with my coworkers,” Get to know my coworkersbetter,” “Be friendly with others,” and “Put myself in others’ shoes.” Thecoefficient alpha was .92 for the star-referenced scale and .93 for theproject-referenced scale.

Supportive relationships. Employees assessed the support they pro-vided to the stars and projects using the seven-item helping scale from Van


Dyne and LePine (1998). The items included “Volunteered to do thingsfor them,” “Helped them with their work responsibilities,” “Helped themlearn about the work,” “Gotten involved to benefit them,” “Attended func-tions that help them,” “Assisted them with their work for the benefit ofthe group,” and “Helped orient new employees for them.” The coefficientalpha was .87 for the supportive relationship with star scale and .85 forthe supportive relationship with project scale.

Supervisor Measures

Positive affect. Following Study 1, positive affect toward the employeewas measured using the PANAS-X (Watson & Clark, 1994). The five-point scale ranged from 1 = not at all to 5 = to a very large extent. Thecoefficient alpha for this scale was .96.

Perceived promotability. As in Study 1, the supervisor’s perception ofthe employee’s overall promotability was measured with Harris et al.’s(2006) seven-item scale. This scale had a coefficient alpha of .91.

Control Variables

We examined the potential importance of many of the same controlsfrom Study 1, including demographic variables and emotional stability(α = .85) using the four-item mini-IPIP scale (Donnellan et al., 2006),again supplemented with four items from the IPIP. None had significanteffects on our hypothesis testing, so they were omitted from our analyses(Carlson & Wu, 2012). The following controls were included, however,to address limitations of Study 1.

Opportunity to observe. Given that supervisors nominated the starsand projects in Study 2, it became important to ensure that employees hadsufficient knowledge of those peers—and to control for any variation insuch knowledge. We therefore assessed employee ratings of the opportu-nity to observe stars and projects using Rodell’s (2013) three-item scale,which was adapted from Judge and Ferris (1993). The items referred theemployee to either the star or project and included “I regularly have theopportunity to observe his/her job performance,” “Most of the time, I amable to monitor his/her job performance,” and “Generally, it is easy forme to see his/her job performance.” The coefficient alpha was .86 for thestar-referenced scale and .87 for the project-referenced scale.

Perceived performance. We controlled for supervisor perceptions ofemployee performance to remove any contaminating influence that theyhad on the positive affect felt by the supervisor for the employee. Per-ceived performance was assessed using a five-item measure adaptedfrom MacKenzie, Podsakoff, and Fetter (1991). The items referred the


supervisor to the employee and included, “Is very good at his/her dailyjob activities,” “Is an excellent worker compared to his/her peers,”“Is outstanding at his/her job, all things considered,” “Is one of the best atwhat he/she does,” and “Is a good performer in general.” The coefficientalpha for this scale was .93.

Perceived citizenship. Similarly, we controlled for supervisor percep-tions of employee citizenship to remove any contaminating influence thatthey had on the positive affect felt by the supervisor for the employee. Per-ceived citizenship was assessed using Lee and Allen’s (2002) eight-itemmeasure. The items referred the supervisor to the employee and included“Attends functions that are not required but that help the organizationalimage,” “Shows pride when representing the organization in public,” “De-fends the organization when other employees criticize it,” “Keeps up withdevelopments in the organization,” “Offers ideas to improve the func-tioning of the organization,” “Takes action to protect the organizationfrom potential problems,” “Expresses loyalty toward the organization,”and “Demonstrates concern about the image of the organization.” Thecoefficient alpha for this scale was .90.

Confirmatory Factor Analysis

We conducted a confirmatory factor analysis in which impressionmanagement motives for stars and projects, prosocial motives for starsand projects, supportive relationships with stars and projects, positiveaffect, perceived promotability, perceived performance, perceived citi-zenship, opportunity to observe star, and opportunity to observe projectwere modeled as latent factors with item-level indicators. Our proposed12-factor model provided a good fit to the data: χ2 (3,849) = 7,566.65,p < .01; CFI = .94; IFI = .94; RMSEA = .067; SRMR = .059. Wecompared our measurement model to an 11-factor model where the per-ceived promotability and positive affect items loaded on a single factor.That comparison revealed a χ2 difference of 466.63 (df = 11, p < .001)in favor of our measurement model. We also compared our measurementmodel to a 10-factor model where the items for perceived performance,positive affect, and perceived citizenship loaded on a single factor. Thatcomparison revealed a χ2 difference of 910.04 (df = 21, p < .001) in favorof our measurement model. Finally, we compared our measurement modelto a nine-factor model where the items for perceived performance, posi-tive affect, perceived citizenship, and perceived promotability loaded on asingle factor. That comparison revealed a χ2 difference of 1,139.55 (df =30, p < .001) in favor of our measurement model. Taken together, theseresults provide support for the structure of our hypothesized measurementmodel.


Results and Discussion

Descriptive Statistics

The descriptive statistics, coefficient alphas, and zero-order correla-tions for the variables in Study 2 are shown in Table 3.

Tests of Hypotheses

We tested the model shown in Figure 2 using structural equation mod-eling in LISREL 8.80. To prevent the number of estimated parametersfrom exceeding our sample size (for a discussion of this issue, see Kline,2011), we created three parcels as indicators for each of the followingconstructs: impression management motives for star and project, proso-cial motives for star and project, supportive relationships with star andproject, positive affect, perceived citizenship, and perceived promotabil-ity. We used all items in the scales, rather than parcels, as indicators foropportunity to observe star, opportunity to observe project, and perceivedperformance, in order to avoid having only one or two parcels per con-struct, which can lead to an underidentified model (Kline, 2011). As inStudy 1, we included direct effects from supportive relationships with starsand projects to perceived promotability because those paths are necessarywhen testing hypotheses of indirect effects (MacKinnon et al., 2002). Weallowed the disturbance terms for positive affect, perceived performance,and perceived citizenship to covary to represent unmeasured commoncauses. The resulting model provided an adequate fit to the data: χ2 (634)= 1,258.90, p < .01; CFI = .96; IFI = .96; RMSEA = .067; SRMR =.071.2 The standardized output from LISREL is shown in Figure 2.

Hypothesis 6a predicted that impression management motives forstars would be positively related to supportive relationships with stars.As shown in Figure 2, the relationship between impression managementmotives for stars and supportive relationships with stars was significant(b = .20). Hypothesis 6a was therefore supported. Figure 2 also revealsa nonsignificant linkage between impression management motives forprojects and supportive relationships with projects. Hypothesis 6b pre-dicted that prosocial motives for projects would be positively related to

2We examined an alternative model where perceived promotability and positive affectwere modeled as dual terminal outcomes, as opposed to positive affect predicting perceivedpromotability. A comparison of the two models revealed a χ 2 difference of 9.97 (df = 1,p < .01) in favor of our hypothesized model. We also examined a model in which perceivedperformance, citizenship, and promotability were all modeled as terminal outcomes. Acomparison of the two models revealed a χ 2 difference of 113.18 (df = 3, p < .01) in favorof our hypothesized model. Thus, we retained our hypothesized model.


TAB

LE

3St

udy

2:D

escr

iptiv

eSt

atis

tics,

Rel

iabi

litie

s,an

dC

orre

latio

ns

Var

iabl

eM

ean

SD1

23

45

67

89

1011

12

1.Im

pres

sion

man

agem

entm

otiv

esfo

rsta

r3.

19.9

5.9

52.

Impr

essi

onm

anag

emen

tmot

ives

forp

roje

ct3.

12.9

3.7

9∗.9

53.

Pros

ocia

lmot

ives

fors

tar

4.03

.57

.26∗

.21∗

.92

4.Pr

osoc

ialm

otiv

esfo

rpro

ject

3.97

.64

.15∗

.27∗

.58∗

.93

5.Su

ppor

tive

rela

tions

hip

with

star

3.85

.64

.28∗

.16∗

.33∗

.22∗

.87

6.Su

ppor

tive

rela

tions

hip

with

proj

ect

3.71

.64

.16∗

.23∗

.20∗

.46∗

.52∗

.85

7.Po

sitiv

eaf

fect

3.98

.78

.16∗

.10

.24∗

.21∗

.23∗

.24∗

.96

8.Pe

rcei

ved

prom

otab

ility

4.26

.63

.13

.04

.20∗

.17∗

.25∗

.23∗

.64∗

.91

9.Pe

rcei

ved

perf

orm

ance

4.34

.62

.01

-.04

.25∗

.24∗

.15∗

.24∗

.58∗

.72∗

.93

10.P

erce

ived

citiz

ensh

ip4.

11.5

9.1

4∗.0

5.2

8∗.2

4∗.2

7∗.2

5∗.6

4∗.6

8∗.6

5∗.9

011

.Opp

ortu

nity

toob

serv

est

ar3.

89.7

3.1

4∗.0

4.4

2∗.2

1∗.4

1∗.1

4∗.1

3.2

2∗.1

6∗.1

8∗.8

612

.Opp

ortu

nity

toob

serv

epr

ojec

t3.

81.7

7.1

5∗.1

5∗.3

1∗.3

1∗.2

6∗.3

3∗.1

5.1

8∗.2

2∗.1

5∗.5

5∗.8

7

Not

e.n

=20

4.C

oeffi

cien

talp

has

are

onth

edi

agon

al.

∗ p<

.05.


Figu

re2:

Stru

ctur

alE

quat

ion

Mod

elin

gR

esul

tsfo

rSt

udy

2.

Not

e.n

=20

4.*p

<.0

5.


TABLE 4Study 2: Effect Decomposition Results for Effects of Supportive Relationships

With Stars and Projects on Perceived Promotability

Predictor

Supportive relationship Supportive relationshipwith star with project

Indirect effect throughPositive affect .04∗ .04∗

Perceived performance .02 .13∗

Perceived citizenship .06∗ .04∗

Total indirect effect .12 .21∗

Direct effect .11∗ -.05Total effect .23∗ .16∗

Note. n = 204.∗p < .05.

supportive relationships with projects. The relationship between proso-cial motives for projects and supportive relationships with projects wassignificant (b = .41), supporting Hypothesis 6b. Figure 2 also reveals anonsignificant linkage between prosocial motives for stars and supportiverelationships with stars.

Hypotheses 2 and 3 predicted that supportive relationships with starsand projects would be positively related to positive affect on the part ofsupervisors. Both hypotheses were tested while controlling for perceivedperformance and perceived citizenship. As shown in Figure 2, both hy-potheses were supported, as supportive relationships with stars (b = .18)and supportive relationships with projects (b = .18) were significantlyrelated to positive affect.

Hypotheses 4 and 5 predicted that supportive relationships with starsand projects would have indirect effects on perceived promotabilitythrough positive affect. These indirect effects and their significance levelswere calculated using the distribution-of-the-product method (MacKin-non et al., 2007) described in Study 1, with perceived performance andperceived citizenship controlled. The results are shown in Table 4. In sup-port of Hypotheses 4 and 5, the results revealed that, when controlling forperceived performance and perceived citizenship, supportive relationshipswith stars (.04) and projects (.04) both had significant indirect effects onperceived promotability through positive affect.

General Discussion

When deciding how to manage their images in the workplace,employees have a number of options—most of which represent


behaviors targeted at supervisors (Bolino et al., 2008; Gardner & Mar-tinko, 1988; Roberts, 2005; Rosenfeld et al., 1995). Our study focusedinstead on the impression management implications of supportive rela-tionships with peers. Although scholars have previously noted the positiveimage implications for employees who build supportive relationships atwork (Allen et al., 1997; Cialdini, 1989; Kilduff & Krackhardt, 1994),many aspects of this phenomenon have remained unaddressed. In partic-ular, it remains unclear whether supportive relationships with both starand project peers are viewed as impression management opportunities. Italso remains unclear how supervisors react to relationships with those twotypes of peers, in terms of evaluations like positive affect and perceivedpromotability.

Taken together, the results of both studies showed that supportive re-lationships with stars and supportive relationships with projects both hadsignificant effects on perceived promotability—a fundamental outcomein impression management contexts (Ferris & Judge, 1991; Harris et al.,2006; Thacker & Wayne, 1995). We had reasoned that supportive em-ployees would be viewed as more promotable because supervisors wouldfeel positive affect toward those employees. Such affect is, after all, aproximal concern when employees manage impressions in the workplace(Bolino et al., 2006; Cialdini, 1989; Ferris et al., 1994; Kacmar & Carlson,1999; Wayne & Ferris, 1990; Wayne & Liden, 1995). Our theorizing wassupported for both relationships with stars and relationships with projects.It may be that both relationships help supervisors protect relevant goalsand identity elements, triggering positive affect (Lazarus, 1991; Novacek& Lazarus, 1990). For example, employees who support stars may con-tribute to supervisor achievement goals while also appealing to supervisoresteem. For their part, employees who support projects may contributeto supervisor stress avoidance goals while also appealing to supervisormoral values and ideals.

Although both supportive relationships with stars and supportive re-lationships with projects were shown to “move the needle” on supervisorimpressions, employees only acted based on impression management mo-tives in the case of stars. Those motives predicted supportive relationshipswith stars in both studies. In contrast, impression management motivesdid not predict supportive relationships with projects in either study. Theresults for stars are understandable given the intuitive benefits that comefrom being connected to a well-regarded peer (Cialdini, 1989, 2001;Dijkstra et al., 2010; Kilduff & Krackhardt, 1994; Leary, 1995; Schlenker,1980). The results for projects, however, point to a missed impressionmanagement opportunity for many employees. The indirect effect ofsupportive relationships with projects on perceived promotability was


similar to the effect for stars, yet those project relationships were notdriven by impression management motives.

Indeed, our Study 2 results showed just how ripe that missed op-portunity could be when it comes to supporting projects. Specifically,our results showed that supportive relationships with projects—and notsupportive relationships with stars—predicted supervisor perceptions ofemployee performance. It may be that supervisors place a premium onthe task relevance of support (Allen, 2007), sensing that employees maybe sharpening their skills, or growing in their capacity to perform theirjobs by helping those who are less proficient (Allen et al., 1997; Kram,1985; Newby & Heide, 1992; Noe et al., 2002; Wagner & Sternberg,1985). In contrast, supervisors may not see any potential for employeesto improve their skills by helping those who essentially do not need help.Both types of supportive relationships contributed to supervisor percep-tions of employee citizenship, which makes sense given the promotiveand discretionary nature of our supportive relationship operationaliza-tions (Bolino, 1999; Bolino et al., 2006; Grant & Mayer, 2009; Rioux &Penner, 2001).

Of course, if employees were not supporting projects out of concernsfor impression management, then why were they supporting them? OurStudy 2 results suggest that supporting projects stands as an opportunityfor exercising prosocial motives. Prosocial motives were a significant pre-dictor of supportive relationships with projects but not stars. In a studyof motives for helping behavior, Grant and Mayer (2009) had shown thatbeing a “good citizen” (i.e., helping) was driven by both being a “good ac-tor” (i.e., impression management motives) and by being a “good soldier”(i.e., prosocial motives). Our results suggest that employees play the roleof good actor with stars while assuming the part of good soldier aroundprojects. That latter result underscores Grant and Gino’s (2010) findingthat prosocial motives become most salient when employees believe thatthey can truly make a difference.

Collectively, our results make important contributions to the literatureon impression management. They illustrate the value in continuing tomove beyond supervisor-targeted actions by expanding the focus to peer-referenced behaviors (Allen et al., 1997; Kilduff & Krackhardt, 1994;Ragins & Scandura, 1999). Indeed, it may be that employees view suchactions as less risky than tactics like self-promotion or ingratiation, whichcan sometimes have negative reputational implications. In addition, ourresults reinforce the wisdom in Cialdini and Richardson’s (1980) notionthat what is done may be less important than with whom it is done.Once the impression management net is cast beyond supervisors, scholarsshould resist the temptation to lump peers together. Finally, our results


revealed important differences in relationships with stars and projects, andaccounting for such nuances would seem critical in future work.

In addition, our findings contribute to the literature on helping. Ourresults suggest that different motives drive employees’ willingness tohelp certain types of coworkers. This is important for researchers at-tempting to isolate the predictors of various helping behaviors (Organ,Podsakoff, & MacKenzie, 2006). Moreover, our results support theoriz-ing that suggests workplace helping behaviors are not purely altruistic(Bolino, 1999). Indeed, employees may also recognize that helping canhave an influence on how they are perceived by supervisors and importantothers.

Suggestions for Future Research

In addition to these contributions, our results point to some valuabledirections for future research. For example, it is important for researchersto continue to identify different motives for predicting supportive rela-tionships with peers. As in the present research, there are certainly other“good actor” and “good citizen” motives that could drive employees to-ward forming workplace relationships with their peers (Bolino, 1999; Kimet al., 2013; Rioux & Penner, 2001). Relatedly, researchers should alsoconsider other operationalizations of supportive relationships, or othertypes of interpersonal connections between employees, as outcomes ofimpression management motives. For example, it could be that socialnetworks have a similar ability to influence supervisor reactions as dosupportive relationships. Results from these studies would be instrumen-tal in identifying employees who are likely to receive help or not andfor aiding organizations with developing unbiased feedback processes forsupervisors.

Our study focused on how supervisors would be influenced by support-ive relationships, creating a need for future studies to consider how otherparties might also be influenced. For example, would projects react nega-tively if they knew that the help they received was for image-enhancementpurposes? Or, would they be grateful for the assistance, regardless of theemployees’ intent? Also, how would observers—or coworkers not directlyinvolved in supportive relationships—react? Would they react positivelybecause they indirectly benefit from their peers’ development or becausethey are not burdened with the responsibility of supporting others? Or, asimpression management scholars have noted, would they be suspiciousof their colleagues whom they perceive as leveraging these relationshipssolely to win the boss’s favor (Vonk, 2002)?

Other suggestions center on additional nuances residing within the“peer” referent. We were drawn to the star versus project distinction


because most work units likely include examples of both and because re-lationships with both types of peers could reveal impression managementbenefits. That said, most work units include other kinds of peers—someof which could be intriguing for their impression management benefits (orcosts). One example could be poor performing peers who lack the devel-opmental potential of projects but who remain employed for idiosyncraticor bureaucratic reasons. How might supervisors view supportive relation-ships with these kinds of peers? As another example, some work unitsinclude peers who engage in minor or major deviant behaviors directed atcoworkers, the organization, or the supervisor (e.g., Bennett & Robinson,2000). Do supervisors appreciate employees giving such peers “secondchances,” or do such connections create reputational detriments for thoseemployees?

Limitations

This research has some limitations that should be noted. One limitationis that our measures of positive affect and perceived promotability werecollected at the same time and from the same source. The path coefficientsin that portion of the models may therefore be inflated by common methodbias (Podsakoff, MacKenzie, Lee, & Podsakoff, 2003). However, it shouldbe noted that the other portions of our models possess some combinationof temporal separation, source separation, or both. Both design elementsrepresent procedural remedies for common method bias (Doty & Glick,1998). Moreover, many of our hypotheses depended on indirect effectsthat were not time or source bound. Another limitation was our smallsample size in Study 1. Our design involved collecting data over two timeperiods and across four sources, which is challenging from an attritionperspective. This issue was exacerbated in our study by the need to identifytwo specific types of peers: stars and projects. Relatedly, the recruitmentapproach used in Study 1 prevented the calculation of a response rate—anissue that was addressed in the design of Study 2. We also note that inStudy 1, our value for SRMR was not ideal, indicating potential modelmisspecification. However, our other fit indices indicated good fit, andthis issue did not arise in Study 2.

Practical Implications

Despite those limitations, our research offers some important practi-cal implications for employees, supervisors, and human resource person-nel. For employees, our results point to important outcomes that can begained from providing support to others. These outcomes include positiveaffect and perceptions that the employee is promotable. Perhaps more


importantly, these gains can come from peers who are at different per-ceived levels of competence and ability. Employees who might otherwisebe tempted to shy away from working with projects (Andrews & Kac-mar, 2001; Cialdini, 1989; Kulik et al., 2008) can be encouraged that itis not only the high-ability peers that offer an image enhancing “returnon investment” at the workplace (Feldman & Klich, 1991). Moreover, ourresults suggest that for employees, providing support might be a viablepath to being recognized as a good citizen or a competent performer atone’s job (Rosenfeld et al., 1995, Wexler, 1986).

For supervisors, our results suggest a compelling story to tell employ-ees when promoting the benefits of a supportive workplace culture (Tews,Michel, & Ellingson, 2013). That story can highlight how employees, inaddition to organizations, might benefit when support is provided to dif-ferent types of peers. The employees’ benefits can come in the form ofpositive supervisor perceptions. However, our results also suggest that su-pervisors should recognize the impact that their perceptions have on howthey evaluate their employees (Bazerman & Moore, 2009). That recog-nition could prompt more extensive record keeping or the collection ofadditional opinions to offset the impact of affect.

Finally, our results could be important to human resource personnel.Most human resource personnel are well aware of the influence that im-pression management can have on the interviewing process and hiringdecisions (Bolino et al., 2008; Rosenfeld et al., 1995). However, less rec-ognized by these personnel is how impression management can impactposthiring decisions and outcomes, such as supervisor evaluations andpromotions. By recognizing that employees might use support as an im-age management strategy, human resource personnel can work to ensurethat support is used with good overall intentions and not merely as ameans to look good. Moreover, human resource personnel can help trainsupervisors to be wary of employees who are using support with dubiousintentions (Halbesleben, Bowler, Bolino, & Turnley, 2010).

REFERENCES

Allen TD. (2004). Protege selection by mentors: Contributing individual and or-ganizational factors. Journal of Vocational Behavior, 65, 469–483. doi:10.1016/j.jvb.2003.07.003

Allen TD. (2007). Mentoring relationships from the perspective of the mentor. In RaginsBR, Kram KE (Eds.), The handbook of mentoring at work: Theory, research andpractice (pp. 123–148). Thousand Oaks, CA: Sage.

Allen TD, Lentz E, Day R. (2006). Career success outcomes associated with mentoringothers: A comparison of mentors and nonmentors. Journal of Career Development,32, 272–285. doi: 10.1177/0894845305282942


Allen TD, Poteet ML, Burroughs SM. (1997). The mentor’s perspective: A qualitativeinquiry and future research agenda. Journal of Vocational Behavior, 51, 70–89. doi:10.1006/jvbe.1997.1596

Allen TD, Rush MC. (1998). The effects of organizational citizenship behavior on perfor-mance judgments: A field study and a laboratory experiment. Journal of AppliedPsychology, 83, 247–260. doi: 10.1037/0021-9010.83.2.247

Andrews MC, Kacmar KM. (2001). Impression management by association: Construc-tion and validation of a scale. Journal of Vocational Behavior, 58, 142–161. doi:10.1006/jvbe.2000.1756

Balzer WK, Sulsky LM. (1992). Halo and performance appraisal research: A criticalexamination. Journal of Applied Psychology, 77, 975–985. doi: 10.1037/0021-9010.77.6.975

Baumeister RF. (1982). A self-presentational view of social phenomena. PsychologicalBulletin, 91, 3–26. doi: 10.1037/0033-2909.91.1.3

Bazerman MH, Moore DA. (2009). Judgment in managerial decision making. Hoboken,NJ: John Wiley & Sons.

Bennett RJ, Robinson SL. (2000). Development of a measure of workplace deviance.Journal of Applied Psychology, 85, 349–360. doi: 10.1037/0021-9010.85.3.349

Bolino MC. (1999). Citizenship and impression management: Good soldiers or good actors?Academy of Management Review, 24, 82–98. doi: 10.5465/AMR.1999.1580442

Bolino MC, Kacmar KM, Turnley WH, Gilstrap JB. (2008). A multi-level review ofimpression management motives and behaviors. Journal of Management, 34, 1080–1109. doi: 10.1177/0149206308324325

Bolino MC, Varela JA, Bande B, Turnley WH. (2006). The impact of impression manage-ment tactics on supervisor ratings of organizational citizenship behavior. Journal ofOrganizational Behavior, 27, 281–297. doi: 10.1002/job.379

Bowler WM, Brass DJ. (2006). Relational correlates of interpersonal citizenship behav-ior: A social network perspective. Journal of Applied Psychology, 91, 70–82. doi:10.1037/0021-9010.91.1.70

Bozeman DP, Kacmar KM. (1997). A cybernetic model of impression management pro-cesses in organizations. Organizational Behavior and Human Decision Processes,69, 9–30. doi: 10.1006/obhd.1996.2669

Bozionelos N. (2004). Mentoring provided: Relation to mentor’s career success, person-ality, and mentoring received. Journal of Vocational Behavior, 64, 24–46. doi:10.1016/S0001-8791(03)00033-2

Cardy RL, Dobbins GH. (1986). Affect and appraisal accuracy: Liking as an integraldimension in evaluating performance. Journal of Applied Psychology, 71, 672–678.doi: 10.1037/0021-9010.71.4.672

Carlson KD, Wu J. (2012). The illusion of statistical control: Control variable practicein management research. Organizational Research Methods, 15, 413–435. doi:10.1177/1094428111428817

Cialdini RB. (1989). Indirect tactics of image management: Beyond basking. In GiacaloneRA, Rosenfeld P (Eds.), Impression management in the organization (pp. 45–56).Hillsdale, NJ: Lawrence Erlbaum Associates.

Cialdini RB. (2001). Influence: Science and practice. Boston, MA: Allyn & Bacon.Cialdini RB, de Nicholas ME. (1989). Self-presentation by association. Journal of Person-

ality and Social Psychology, 57, 626–631. doi: 10.1037/0022-3514.57.4.626Cialdini RB, Richardson KD. (1980). Two indirect tactics of image management: Bask-

ing and blasting. Journal of Personality and Social Psychology, 39, 406–415. doi:10.1037/0022-3514.39.3.406


Colquitt JA, Baer MD, Long DM, Halvorsen-Ganepola MDK. (2014). Scale indicators ofsocial exchange relationships: A comparison of relative content validity. Journal ofApplied Psychology, 99, 599–618. doi: 10.1037/a0036374

Crandall CS, Silvia PJ, N’Gbala AN, Tsang JA, Dawson K. (2007). Balance theory, unitrelations, and attribution: The underlying integrity of Heiderian theory. Review ofGeneral Psychology, 11, 12–30. doi: 10.1037/1089–2680.11.1.12

DeNisi AS, Williams KJ. (1988). Cognitive approaches to performance appraisal. In FerrisGR, Rowland KM (Eds.), Research in personnel and human resources management(Vol. 6, pp. 109–155). Greenwich, CT: JAI Press.

Dijkstra JK, Cillessen AH, Lindenberg S, Veenstra R. (2010). Basking in reflected gloryand its limits: Why adolescents hang out with popular peers. Journal of Researchon Adolescence, 20, 942–958. doi: 10.1111/j.1532-7795.2010.00671.x

Donnellan MB, Oswald FL, Baird BM, Lucas RE. (2006). The mini-IPIP scales: Tiny-yet-effective measures of the Big Five factors of personality. Psychological Assessment,18, 192–203. doi: 10.1037/1040-3590.18.2.192

Doty DH, Glick WH. (1998). Common methods bias: Does common methods vari-ance really bias results? Organizational Research Methods, 1, 374–406. doi:10.1177/109442819814002

Dreher GF, Ash RA. (1990). A comparative study of mentoring among men and women inmanagerial, professional, and technical positions. Journal of Applied Psychology,75, 539–546. doi: 10.1037/0021-9010.75.5.539

DuBrin AJ. (2011). Impression management in the workplace: Research, theory, and prac-tice. New York, NY: Routledge.

Eastman KK. (1994). In the eyes of the beholder: An attributional approach to ingratia-tion and organizational citizenship behavior. Academy of Management Journal, 37,1379–1391. doi: 10.2307/256678

Feldman DC, Klich NR. (1991). Impression management and career strategies. InGiacalone RA, Rosenfeld P (Eds.), Applied impression management: How imagemaking affects organizations (pp. 67–80). Newbury Park, CA: Sage.

Ferris GR, Hochwarter WA, Douglas C, Blass FR, Kolodinsky RW, Treadway DC. (2002).Social influence processes in organizations and human resources systems. In FerrisGR, Martocchio JJ (Eds.), Research in personnel and human resources management(Vol. 21, pp. 65–127). Oxford, UK: JAI Press/Elsevier Science. doi: 10.1016/S0742-7301(02)21002-6

Ferris GR, Judge TA. (1991). Personnel/human resources management: A politi-cal influence perspective. Journal of Management, 17, 447–488. doi:10.1177/014920639101700208

Ferris GR, Judge TA, Rowland KM, Fitzgibbons DE. (1994). Subordinate influence andthe performance evaluation process: Test of a model. Organizational Behavior andHuman Decision Processes, 58, 101–135. doi: 10.1006/obhd.1994.1030

Floyd SW, Wooldridge B. (1994). Dinosaurs or dynamos? Recognizing middle man-agement’s strategic role. Academy of Management Executive, 8, 47–57. doi:10.5465/AME.1994.9412071702

Frijda NH. (1994). Varieties of affect: Emotions and episodes, moods, and affect. In EkmanP, Davidson RJ (Eds.), The nature of emotion (pp. 59–67). New York, NY: OxfordUniversity Press.

Gardner W, Martinko MJ. (1988). Impression management in organizations. Journal ofManagement, 14, 321–338. doi: 10.1177/014920638801400210

Gardner W, Martinko MJ. (1998). An organizational perspective of the effects of dysfunc-tional impression management. In Griffin RW, O’Leary-Kelly A, Collins JM (Eds.),


Dysfunctional behavior in organizations: Non-violent dysfunctional behavior (pp.69–125). Stamford, CT: JAI Press.

Grant AM. (2008). Does intrinsic motivation fuel the prosocial fire? Motivational syn-ergy in predicting persistence, performance, and productivity. Journal of AppliedPsychology, 93, 48–58. doi: 10.1037/0021-9010.93.1.48

Grant AM, Gino F. (2010). A little thanks goes a long way: Explaining why gratitude ex-pressions motivate prosocial behavior. Journal of Personality and Social Psychology,98, 946–955. doi: 10.1037/a0017935

Grant AM, Mayer DM. (2009). Good soldiers and good actors: Prosocial and impressionmanagement motives as interactive predictors of affiliative citizenship behaviors.Journal of Applied Psychology, 94, 900–912. doi: 10.1037/a0013770

Halbesleben JRB, Bowler WM, Bolino MC, Turnley WH. (2010). Organizational concern,prosocial values, or impression management? How supervisors attribute motivesto organizational citizenship behavior. Journal of Applied Social Psychology, 40,1450–1489. doi: 10.1111/j.1559-1816.2010.00625.x

Ham J, Vonk R. (2011). Impressions of impression management: Evidence of spontaneoussuspicion of ulterior motivation. Journal of Experimental Social Psychology, 47,466–471. doi: 10.1016/j.jesp.2010.12.008

Harris KJ, Kacmar KM, Carlson DS. (2006). An examination of temporal variables andrelationship quality on promotability ratings. Group & Organization Management,31, 677–699. doi: 10.1177/1059601106286889

Higgins CA, Judge TA, Ferris GR. (2003). Influence tactics and work outcomes: A meta-analysis. Journal of Organizational Behavior, 24, 89–106. doi 10.1002/job.181

Hinkin TR. (1998). A brief tutorial on the development of measures for use insurvey questionnaires. Organizational Research Methods, 1, 104–121. doi:10.1177/109442819800100106

Hinkin TR, Tracey JB. (1999). An analysis of variance approach to content validation.Organizational Research Methods, 2, 175–186. doi: 10.1177/109442819922004

Hunt DM, Michael C. (1983). Mentorship: A career training and development tool. Academyof Management Review, 8, 475–485. doi: 10.5465/AMR.1983.4284603

Isen AM, Shalker T, Clark M, Karp L. (1978). Affect, accessibility of material in memoryand behavior: A cognitive loop? Journal of Personality and Social Psychology, 36,1–12. doi: 10.1037/0022-3514.36.1.1

Jones EE. (1990). Interpersonal perception. New York, NY: Freeman.Jones EE, Pittman T. (1982). Toward a general theory of strategic self-presentation. In

Suls J (Ed.), Psychological perspectives on the self (pp. 231–262). Hillsdale, NJ:Erlbaum.

Joreskog K, Sorbom D. (1989). LISREL 7: A guide to the program and applications.Lincolnwood, IL: Scientific Software International.

Judge TA, Ferris GR. (1993). Social context of performance evaluation decisions. Academyof Management Journal, 36, 80–105. doi: 10.2307/256513

Kacmar KM, Carlson DS. (1999). Effectiveness of impression management tactics acrosshuman resource situations. Journal of Applied Social Psychology, 26, 1293–1315.doi: 10.1111/j.1559-1816.1999.tb02040.x

Kilduff M, Krackhardt D. (1994). Bringing the individual back in: A structural analysisof the internal market for reputation in organizations. Academy of ManagementJournal, 37, 87–108. doi: 10.2307/256771

Kim YJ, Van Dyne L, Kamdar D, Johnson RE. (2013). Why and when do motives matter?An integrative model of motives, role cognitions, and social support as predictorsof OCB. Organizational Behavior and Human Decision Processes, 121, 231–245.doi: 10.1016/j.obhdp.2013.03.004


Kline RB. (2011). Principles and practice of structural equation modeling (3rd ed.). NewYork, NY: Guilford Press.

Kram KE. (1985). Mentoring at work: Developmental relationships in organizational life.Glenview, IL: Scott Foresman.

Kram KE, Isabella LA. (1985). Mentoring alternatives: The role of peer relationshipsin career development. Academy of Management Journal, 28, 110–132. doi:10.2307/256064

Kulik CT, Bainbridge HT, Cregan C. (2008). Known by the company we keep: Stigma-by-association effects in the workplace. Academy of Management Review, 33, 216–230.doi: 10.5465/AMR.2008.27752765

Lazarus RS. (1991). Emotions and adaptation. New York, NY: Oxford University.Leary MR. (1995). Self-presentation: Impression management and interpersonal behavior.

Madison, WI: Brown & Benchmark.Leary MR, Kowalski RM. (1990). Impression management: A literature review and

two-component model. Psychological Bulletin, 107, 34–47. doi: 10.1037/0033-2909.107.1.34

Lee K, Allen NJ. (2002). Organizational citizenship behavior and workplace deviance:The role of affect and cognitions. Journal of Applied Psychology, 87, 131–142. doi:10.1037/0021-9010.87.1.131

Lewis RE, Heckman RJ. (2006). Talent management: A critical review. Human ResourceManagement Review, 16, 139–154. doi: 10.1016/j.hrmr.2006.03.001

MacKenzie SB, Podsakoff PM, Fetter R. (1991). Organizational citizenship behavior andobjective productivity as determinants of managerial evaluations of salespersons’performance. Organizational Behavior and Human Decision Processes, 50, 123–150. doi: 10.1016/0749-5978(91)90037-T

MacKinnon DP, Fritz MS, Williams J, Lockwood CM. (2007). Distribution of the productconfidence limits for the indirect effect: Program PRODCLIN. Behavioral ResearchMethods, 39, 384–389. doi: 10.3758/BF03193007

MacKinnon DP, Lockwood CM, Hoffman JM, West SG, Sheets V. (2002). A comparisonof methods to test mediation and other intervening variable effects. PsychologicalMethods, 7, 83–104. doi: 10.1037/1082-989X.7.1.83

Newby TJ, Heide A. (1992). The value of mentoring. Performance Improvement Quarterly,5, 2–15. doi: 10.1111/j.1937-8327.1992.tb00562.x

Noe RA. (2002). Employee training and development. New York: McGraw-Hill.Noe RA, Greenberger DB, Wang S. (2002). Mentoring: What we know and where we might

go. In Ferris GR, Martocchio JJ (Eds.), Research in personnel and human resourcesmanagement (Vol. 21, pp. 129–173). Oxford, UK: JAI Press/Elsevier Science. doi:10.1016/S0742-7301(02)21003-8

Novacek J, Lazarus RS. (1990). The structure of personal commitments. Journal of Per-sonality, 58, 693–715. doi: 10.1111/j.1467-6494.1990.tb00250.x

Oldroyd JB, Morris SS. (2012). Catching falling stars: A human resource response to socialcapital’s detrimental effect of information overload on star employees. Academy ofManagement Review, 37, 396–418. doi: 10.5465/amr.2010.0403

Organ DW, Podsakoff PM, MacKenzie SB. (2006). Organizational citizenship behavior:Its nature, antecedents, and consequences. Thousand Oaks, CA: Sage.

Podsakoff PM, MacKenzie SB, Lee J, Podsakoff NP. (2003). Common method biases inbehavioral research: A critical review of the literature and recommended remedies.Journal of Applied Psychology, 88, 879–903. doi: 10.1037/0021–9010.88.5.879

Porter COLH, Hollenbeck JR, Ilgen DR, Ellis APJ, West BJ, Moon H. (2003) Backingup behaviors in teams: The role of personality and legitimacy of need. Journal ofApplied Psychology, 88, 391–403. doi: 10.1037/0021-9010.88.3.391


Project. (n.d.). The American Heritage Dictionary of the English Language (4th ed.).Retrieved from http://dictionary.reference.com/browse/project

Ragins BR, Scandura TA. (1999). Burden or blessing? Expected costs and benefits of being amentor. Journal of Organizational Behavior, 20, 493–509. doi: 10.1002/(SICI)1099-1379(199907)20:4<493::AID-JOB894>3.0.CO;2-T

Richardson KD, Cialdini RB. (1981). Basking and blasting: Tactics of indirect self-presentation. In Tedeschi JT (Ed.), Impression management theory and social psy-chological research (pp. 41–53). New York, NY: Academic Press.

Rioux SM, Penner LA. (2001). The causes of organizational citizenship behavior: A motiva-tional analysis. Journal of Applied Psychology, 86, 1306–1314. doi: 10.1037/0021-9010.86.6.1306

Robbins TL, DeNisi AS. (1994). A closer look at interpersonal affect as a distinct influenceon cognitive processing in performance evaluations. Journal of Applied Psychology,79, 341–353. doi: 10.1037/0021-9010.79.3.341

Roberts LM. (2005). Changing faces: Professional image construction in diverseorganizational settings. Academy of Management Review, 30, 685–711. doi:10.5465/AMR.2005.18378873

Rodell JB. (2013). Finding meaning through volunteering: Why do employees volunteerand what does it mean for their jobs? Academy of Management Journal, 56, 1274–1294. doi: 10.5465/amj.2012.0611

Rosenfeld P, Giacalone RA, Riordan CA. (1995). Impression management in organizations:Theory, measurement, practice. London: Routledge.

Schlenker BR. (1980). Impression management: The self-concept, social identity, andinterpersonal relations. Monterey, CA: Brooks/Cole Publishing Company.

Schriesheim CA, Powers KJ, Scandura TA, Gardiner CC, Lankau MJ. (1993). Im-proving construct measurement in management research: Comments and aquantitative approach for assessing the theoretical adequacy of paper-and-pencil and survey-type instruments. Journal of Management, 19, 385–417. doi:10.1177/014920639301900208

Shadish WR, Cook TD, Campbell DT. (2002). Experimental and quasi-experimental de-signs for generalized causal inference. Boston, MA: Houghton Mifflin.

Shaughnessy BA, Treadway DC, Breland JA, Williams LV, Brouer RL. (2011). Influenceand promotability: The importance of female political skill. Journal of ManagerialPsychology, 26, 584–603. doi: 10.1108/02683941111164490

Sherony KM, Green SG. (2002). Coworker exchange: Relationships between coworkers,leader-member exchange, and work attitudes. Journal of Applied Psychology, 87,542–548. doi: 10.1037/0021-9010.87.3.542

Tews MJ, Michel JW, Ellingson JE. (2013). The impact of coworker support on employeeturnover in the hospitality industry. Group & Organization Management, 38, 630–653. doi: 10.1177/1059601113503039

Thacker RA, Wayne SJ. (1995). An examination of the relationship between upward influ-ence tactics and assessments of promotability. Journal of Management, 21, 739–756.doi: 10.1177/014920639502100408

Tofighi D, MacKinnon D. (2011). RMediation: An R package for mediation analysis con-fidence intervals. Behavior Research Methods, 43, 692–700. doi: 10.3758/s13428-011-0076-x

Tsui AS, Barry B. (1986). Interpersonal affect and rating errors. Academy of ManagementJournal, 29, 586–599. doi: 10.2307/256225

Van Dyne L, LePine JA. (1998). Helping and voice extra-role behaviors: Evidence ofconstruct and predictive validity. Academy of Management Journal, 41, 108–119.doi: 10.2307/256902


Vonk R. (2002). Self-serving interpretations of flattery: Why ingratiation works. Journal ofPersonality and Social Psychology, 82, 515–526. doi: 10.1037/0022-3514.82.4.515

Wagner RK, Sternberg RJ. (1985). Practical intelligence in real-world pursuits: The role oftacit knowledge. Journal of Personality and Social Psychology, 49, 436–458. doi:10.1037/0022-3514.49.2.436

Wasserman J. (2014). Ranking the biggest project picks in the 2014 NBA draft class.Bleacher Report, http://bleacherreport.com/articles/1931752-ranking-the-biggest-project-picks-in-the-2014-nba-draft-class.

Watson D, Clark LA. (1994). The PANAS-X: Manual for the positive and negative affectschedule – expanded form. Unpublished manuscript, University of Iowa.

Watson D, Clark LA, Tellegen A. (1988). Development and validation of brief measures ofpositive and negative affect: The PANAS scales. Journal of Personality and SocialPsychology, 54, 1063–1070. doi: 10.1037/0022-3514.54.6.1063

Wayne SJ, Ferris GR. (1990). Influence tactics, affect, and exchange quality in supervisor-subordinate interactions: A laboratory experiment and field study. Journal of AppliedPsychology, 75, 487–499. doi: 10.1037/0021-9010.75.5.487

Wayne SJ, Liden RC. (1995). Effects of impression management on performance rat-ings: A longitudinal study. Academy of Management Journal, 38, 232–260. doi:10.2307/256734

Wexler MN. (1986). Impression management and the new competence: Conjectures forseekers. Et cetera, 43, 247–258.

Yun S, Takeuchi R, Liu W. (2007). Employee self-enhancement motives and job perfor-mance behaviors: Investigating the moderating effects of employee role ambiguityand managerial perceptions of employee commitment. Journal of Applied Psychol-ogy, 92, 745–756. doi: 10.1037/0021-9010.92.3.745

WHAT WILL THE BOSS THINK? THE IMPRESSION...

Documents

Transcript of WHAT WILL THE BOSS THINK? THE IMPRESSION...