xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 2200385Ka 4500
controlfield tag 001 002029120
007 cr bnu|||uuuuu
008 090916s2009 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002893
Taing, Meng Uoy.
Employee commitment :
b the combined effects of bases and foci
h [electronic resource] /
by Meng Uoy Taing.
[Tampa, Fla] :
University of South Florida,
Title from PDF of title page.
Document formatted into pages; contains 68 pages.
Thesis (M.A.)--University of South Florida, 2009.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
ABSTRACT: Recent studies indicate that employees distinguish between commitments to interpersonal foci within the organization, such as supervisors and coworkers. Often, these commitments account for variance in outcomes incremental to organizational commitment (e.g., Becker, 1992). Unfortunately, research has tended to focus on affective forms of commitment to foci, while ignoring normative and continuance commitment. To address this gap, the current study proposed and tested models of commitment to foci which incorporate normative and continuance commitment in addition to affective commitment. Results showed some parallels with findings concerning organizational commitment. Much like organizational commitment, support from a focus relates to affective commitment to that focus, while expectations from a focus predict normative commitment to the focus. Additionally, both affective and normative commitment to supervisors and coworkers predicted favorable outcomes, but continuance commitment did not. In line with researchers recommendations (e.g., Johnson, Groff & Taing, in press), interactions among different bases and foci of commitment were also examined. Exploratory analyses suggested a three-way interaction between affective organizational, supervisor, and coworker commitment for predicting in-role performance.
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
Advisor: Russell E. Johnson, Ph.D.
t USF Electronic Theses and Dissertations.
Employee Commitment: The Combined Effects of Bases and Foci by Meng Uoy Taing A thesis submitted in partial fulfillment of the requirements for the degree of Master of Arts Department of Psychology College of Arts and Sciences University of South Florida Major Professor: Russell E. Johnson, Ph.D Chu-Hsiang Chang, Ph.D. Stephen Stark, Ph.D. Jennifer Bosson, Ph.D. Date of Approval: Mar 24, 2009 Keywords: Supervisor Commitment, Cowork er Commitment, Mult iple Commitments, Commitment Foci, Organizational Commitment, Three-component model Copyright 2009, Meng Uoy Taing
Dedication Many important people in my life inspired th e writing of this th esis. I would like to dedicate this work to my wife, Leakhena. She encouraged me to try my hardest and not to lose sight of my goals. I would also like to acknowledge the support provided by my mother Yon, father Nam, older brothers H eang and Pheng, and older sister Houng. It is with great pride that I dedicate this work to them.
Acknowledgments I would like to acknowledge the professiona l assistance provided by committee members, Daisy Chang, Stephen Stark, and Jennifer Bo sson. They provided essential feedback and advice about how to make my thesis a true contribution to the fi eld of industrial and organizational psychology. I al so extend a special thanks to my major professor, Russ Johnson. He has played a major role in my professional development and provided invaluable assistance with this thesis and other projects. I thank him for his continued encouragement and mentorship.
i Table of Contents List of Tables ii List of Figures iii Abstract iv Chapter OneIntroduction 1 Organizational Commitment 3 Commitment to Interpersonal Foci 6 A Structural Model of Supervisor Commitment 8 A Structural Model of Coworker Commitment 12 Interactive Effects of Commitme nt to Interpersonal Foci 15 Interactive Effects of Commit ment across Different Foci 18 Chapter Two-Method 22 Participants and Procedure 22 Measures 23 Chapter ThreeResults 27 Structural Model of Supervisor Commitment 31 Structural Model of Coworker Commitment 34 Interactive Effects among Ba ses within a Single Focus 37 Interactive Effects of Affective Commitment across Different Foci 39 Exploratory Analyses 42 Chapter FourDiscussion 44 Implications and Directions for Future Research 47 Limitations 49 Conclusion 51 References 52 Appendices 60 Appendix A: Survey Measures 61
ii List of Tables Table 1. Descriptive St atistics and Co rrelations Among Study Variables 28 Table 2. Results of Confirmatory Factor Analysis of Commitment Variables 31 Table 3. Results for Supervisor Model Path Coefficients 34 Table 4. Results for Coworker Model Path Coefficients 36 Table 5. Regression Results for Supervis or Commitment Base Interactions 38 Table 6. Regression Results for Cowork er Commitment Base Interactions 39 Table 7. Regression Results for the Inte raction between Aff ective Supervisor 40 and Coworker Commitment for Predicting Citizenship Behaviors Directed toward Individuals Table 8. Regression Results for Tests of 3-way Interactio ns across Foci 40 Table 9. Summary of Hypotheses and Results 41
iii List of Figures Figure 1. Structural Model of Supervisor Commitment 12 Figure 2. Structural Model of Coworker Commitment 15 Figure 3. Compensatory Interaction Model 16 Figure 4. Synergistic Interaction Model 17 Figure 5. Nested Confirmatory Factor Analytic Models 30 Figure 6. Standardized Estimates fo r Supervisor Structural Model 33 Figure 7. Standardized Estimates for Coworker Structural Model 36 Figure 8. Three-Way Interaction between Affective Commitments to Foci 43 for Predicting In-role Performance
iv Employee Commitment: The Combined Effects of Bases and Foci Meng Uoy Taing ABSTRACT Recent studies indicate that employees distinguish between commitments to interpersonal foci within the organization, such as supervisors and coworkers. Often, these commitments account for variance in outcomes incremental to organizational commitment (e.g., Becker, 1992). Unfortunate ly, research has tended to focus on affective forms of commitment to foci, while ignoring normative and continuance commitment. To address this gap, the curre nt study proposed and tested models of commitment to foci which incorporate normative and continuance commitment in addition to affective commitment. Results showed some parallels with findings concerning organizational commitment. Much like organizational commitment, support from a focus relates to affective commitment to that focus, while expectations from a focus predict normative commitment to th e focus. Additionally, both affective and normative commitment to supervisors and co workers predicted favorable outcomes, but continuance commitment did not. In line with researchers recommendations (e.g., Johnson, Groff & Taing, in press), interac tions among different bases and foci of commitment were also examin ed. Exploratory analyses sugg ested a three-way interaction between affective organizationa l, supervisor, and coworker commitment for predicting in-role performance.
1 Chapter One Introduction Recent research has demonstrated the importance of differentiating between commitments to interpersonal foci within th e organization, such as supervisors (Becker, 1992; Becker & Billings, 1993; Becker, Billings Eveleth, & Gilbert, 1996; Stinglhamber, Bentein, & Vandenberghe, 2002; Vandenberghe Bentein, & Stinglhamber, 2004) and coworkers (Bryant, 2001; Wasti & Can, 2008) These studies reveal that not only do employees distinguish between commitments to such foci (Snape, Chan, & Redman, 2006; Bishop, Scott, Goldsby, & Cropanzano, 2005), but often they account for variance in outcomes incremental to organizational commitment (Becker, 1992; Becker & Kernan, 2003; Stinglhamber et al., 2002). Like or ganizational commitment, the mindset accompanying commitment to a particular focus can be characterized as involving affective, normative, and continuance base s (Stinglhamber et al., 2002). As such, it is surprising that most of the re search concerning commitment to foci has focused solely on the affective base. However, doing so may be problematic for several reasons. First, because different motivational mindsets acco mpany each base of commitment (Meyer, Becker, & Vandenberghe, 2004), it is likely that particular bases have unique antecedents and varying effects on outcomes. Secondly, a growing body of evidence suggests that interactions exist among the bases of or ganizational commitment (e.g., Jaros, 1997; Johnson, Groff, & Taing, in press; Somers, 1995), which raises the possibility that
2 interactions also characterize commitment to interpersonal foci. If so, measuring only affective commitment to foci and ignoring potential interactions among bases can result in model misspecification. Of equal importance is the fact that employees can be simultaneously committed to multiple foci within the organization. Re search suggests that employees engage in distinct exchange relationships with multiple organizational constitu encies (Bishop et al., 2005; Stinglhamber & Vandenberghe, 2003; Va ndenberghe et al., 2004) yet few studies have investigated the interactive effects of being committed to more than one focus. Snape, Chan, and Redman (2006) examined interactions in a Chinese sample. However, their results may not generalize to a We stern population because the nature of commitment in Chinese contexts is thought to differ from other cultures (Chen, Tsui, & Farh, 2002; Cheng, Jiang, and Riley, 2003; Farh, Early, & Lin, 1997). Becker and Billings (1993) studied the combined effects of commitment to the organization, supervisor, workgroup, and top management. Based on their patterns of commitment, employees were classified as having a particular Â“commitment profile.Â” Although a profile approach makes it easier to interpret the effect of commitment to multiple foci, a great deal of precision is lost through artificial categorizati on. Furthermore, their analyses confounded additive effects with interactive ones. Finally, Vandenberghe and Bentein (in press) examined interactive effects between affective or ganizational and supervisor commitment. However, they only explored interaction effects on turnover variables, rather than also considering variables such as performance. As a whole, these studies suggest that interactive e ffects characterize commitment to multiple constituencies.
3 However, a full understanding of th e nature of these interactions is far from complete. The purpose of the current study is threefold. First, it fills gaps in the literature by examining the three bases of commitment (i.e ., affective, normative, and continuance) to supervisors and coworkers. Such investigation is needed as it is likely that each base has different antecedents and explains unique vari ance in the prediction of outcomes. Further, it provides an important opportunity to test whether relationships concerning bases of organizational commitment can be generalized to bases of commitment to foci. Second, the current study examines possible interac tive effects among the bases of commitment within each focus. To the au thorÂ’s knowledge, no study thus fa r has explored this issue. Doing so is necessary because although affective commitment has generally been shown to have positive effects on outcomes, if in teractions exist the effects of affective commitment may depend on the relative levels of normative and continuance commitment. Finally, this study considers inte ractions between commitments to different foci. Since employees show varying leve ls of commitment to the organization, supervisor, and coworkers, it is worthwhile to determine whether the combined effects of such commitments differ from their main effects. Thus far, research suggests that interactions between commitmen ts exist, but our understanding of thes e effects is still preliminary. Organizational Commitment Perhaps the most widely studied type of employee commitment is organizational commitment, which is defined as a psychologi cal force that binds employees to their organization and makes turnover less likely (Allen & Meyer, 1990). Organizational
4 commitment is commonly conceptualized as consisting of three distinguishable bases: affective organizational commitment (AOC ), normative organizational commitment (NOC), and continuance organizational co mmitment (COC; Allen & Meyer, 1990). AOC involves an emotional attachment to, involvement in, and identification with oneÂ’s organization (Meyer & Allen, 1991). AOC arises from positive social exchanges between the employee and organization, wh ich are based on perceptions of support (Eisenberger, Huntington, Hutchison, & Sowa, 1986; Shore, Tetrick, Lynch, & Barksdale, 2006) and fairness (Cohen-Ch arash & Spector, 2001; Colquitt, Conlon, Wesson, Porter, & Ng, 2001). NOC is based on a perceived obligation to maintain membership in the organization, which is grounded in a sense of morality. NOC is thought to result primarily from early socialization e xperiences or as a form of reciprocity for organizational benefits (Meyer & Alle n, 1991; Powell & Meyer, 2004). Lastly, COC is derived from the perceived costs of leav ing the organization, in cluding the loss of investments and difficulty in findi ng a new job (Meyer & Allen, 1984). AOC tends to have the strongest relati onship with desirable outcomes (Meyer, Stanley, Herscovitch, & Topolnytsky, 2002). NOC al so tends to relate favorably, but to a lesser degree than AOC (Meyer et al., 2002) With the exception of turnover and turnover intentions, COC tends to be unrelated or unfavorably related to outcomes (Mathieu & Zajac, 1990; Meyer et al., 2002). That each ba se demonstrates relationships of varying strength with outcomes can perhaps be e xplained by the motivational mindsets which underlie each type of commitment (Meyer et al., 2004). Those with high AOC perceive congruence between their goals and those of the organization, which leads to
5 organizational goals being intrinsically and au tonomously regulated (Meyer et al., 2004). On the other hand, NOC reflects commitment based on a moral obligation to remain, which is likely associated with introjected regulation (Gagn & Deci 2005; Meyer et al., 2004). Introjected regulation re flects a weak form of aut onomous motivation in which behaviors are performed to avoid feeli ngs of guilt and shame (Deci & Ryan, 1985). Finally, because COC involves commitment based on external costs, it is thought to be associated with external regulation, the leas t autonomous form of motivation (Meyer et al., 2004). According to Deci and Ryan ( 1985), effort and performance are at their highest when people operate based on intrinsic or autonomous motivation. This proposition has been supported by a number of studies (e.g., Gr olnick & Ryan, 1987; Ryan & Connell, 1989). Meyer and Herscovitch (2001) state that the effects of each base of commitment also depend on whether an outcome is consid ered focal or discretionary. Focal outcomes are attitudes and behaviors that are, by defi nition, implied by the commitment (Meyer et al., 2004; Meyer & Herscovitch, 2001). Discretio nary attitudes and behaviors are ones that arenÂ’t necessarily implied by the commit ment, but may be influenced by it (Meyer & Herscovitch, 2001). For example, organizati onal commitment implies that an employee will stay with the organization, but it does not require that an employee performs organizational citizenship behaviors (OCB), si nce they are not clearly stated to be conditions for employment (Borman & Motowidlo, 1993; Organ & Ryan, 1995). According to Meyer and Herscovitch (2001), all three bases of commitment imply a greater inclination to remain with the organization because staying is focal to
6 organizational commitment. However, onl y AOC and NOC should necessarily have positive effects for discretionary outcomes because they represent mindsets (i.e. desire and obligation to remain, respectively) which involve some level of concern for the wellbeing of the organization. High levels of CO C involve the perception that the costs of leaving the organization are great and do not imply any desire to do more for the organization than the bare minimum of main taining membership (Gellatly, Meyer, & Luchak, 2006). Commitment to Interpersonal Foci Reichers (1985) argued that commitment to the organization may involve multiple constituencies. That is, the organization is an abstraction represented in reality by supervisors, coworkers, and other individuals related to the organi zation. Past research has shown that distinguishing between foci of commitment is useful in that commitment to different foci account for variance in cremental to organizational commitment (e.g., Becker, 1992; Becker & Kernan, 2003; Stingl hamber et al., 2002). For some outcomes, they have even been shown to relate more closely (e.g., Cheng et al., 2003; Vandenberghe et al, 2004, Vandenberghe, Be ntein, Michon, Chebat, Tremblay, & Fils, 2007). Quite often, these stronger relationships with outcomes have been explained by the Â“compatibility hypothesisÂ” (Cheng et al., 2003) or Â“salience of behaviorÂ” (Vandenberghe et al., 2004), which states th at commitment to a focus (e.g., oneÂ’s work team) is a better predictor of behavior toward that focus (e.g., team cohesion) than commitment to a less relevant target such as the organization as a whole. On the other
7 hand, when predicting an organization-relevant outcome such as turnover, organizational commitment will be most influential. This id ea is often credited to Ajzen and FishbeinÂ’s (1977) principle of compatibility, which states that an attitude w ill predict a behavior only to the extent that the attitude is relate d to the behavior. LewinÂ’ s (1943) field theory, which contends that behavior is most st rongly influenced by the elements in the environment that are perceived as being most salient or proximal, is also widely cited. Like organizational commitment, commitmen t to interpersonal foci encompasses the dimensions of affective, normative, and continuance commitment (Stinglhamber et al., 2002). Unfortunately, few studies have exam ined the effects of non-affective forms of commitment to interpersonal foci. A notable exception is Stinglhamber, Bentein, and Vandenberghe (2002), who developed scales for measuring affective, normative, and continuance commitment to five different fo ci. They showed that employees, in fact, distinguish between the bases of commitment to each focus. Additionally, Becker and Kernan (2003) explored the effects of aff ective and continuance (but not normative) supervisor commitment on in-role performa nce and OCB, finding evidence for stronger influences of affective supervisor commitment than continuance supervisor commitment on outcomes. Most recently, Wasti and Can (2008) showed that normative supervisor commitment accounted for variance in job stre ss and OCB directed toward the supervisor incremental to NOC. They did not, however, include affective and normative forms of commitment in the same regression model. Clearly, more research is needed to determine the effects normative and continua nce commitment to interpersonal foci.
8 A Structural Model of Supervisor Commitment Research on commitment to the superv isor has focused almost exclusively on affective supervisor commitment (ASC). This research has shown that, while organizational commitment can arise from perceptions of or ganizational support (Eisenberger et al., 1986; Shore et al., 2006) ASC can result from perceptions of positive leader-member exchanges (LMX; Vandenbe rghe et al., 2004). Although not much research has explored potential antece dents of normative (NSC) and continuance supervisor commitment (CSC), findings from organizational commitment provide a basis for making predictions. It is thought that NOC derives from socialization experiences about proper behavior, such as the need to r eciprocate favors and to live up to othersÂ’ expectations (Powell & Meyer, 2004). As such, it stands to reason that NSC should be influenced by oneÂ’s perceptions surrounding su pervisor expectations about staying with the organization. Additionally, it has been argued that perceived support from the organization creates an obligation for an employee to reciprocate by giving the organization his or her affective commitmen t (Rhoades & Eisenberger, 2002). Wasti and Can (2008) posited that this moral obliga tion to give commitment should involve normative commitment in addition to affectiv e commitment. Indeed, they showed that perceptions of employee empowerment, which is primarily implemented by the supervisor, had positive implications for bot h ASC and NSC. Based on this rationale, we would expect that, in addition to ASC, NSC can arise from perceptions of positive LMX. COC is thought to result from the accum ulation of side-bets which increase the cost of leaving the organi zation over time (Meyer & Alle n, 1984; Meyer & Allen, 1991).
9 These side-bets can include monetary benefits such as bonuses, or other investments such as the acquisition of non-transf errable skills (labeled Â“indivi dual adjustments to social positionsÂ” by Powell & Meyer, 2004). Individual adjustments may be particularly relevant to the development of CSC because staying with the same supervisor for an extended period of time may involve learning ski lls and procedures that are only relevant to working with that supervisor. Changing supervisors may necessitate new training or the need to adjust to supervisor expectations Therefore, it is possibl e that perceptions of high individual adjustments pr edict higher levels of CSC. Based on the reasoning above, the following hypotheses are put forth: Hypothesis 1: LMX is positively related to a) ASC and b) NSC. Hypothesis 2: EmployeesÂ’ perceived exp ectations from the supervisor about staying in the organization ar e positively related to NSC. Hypothesis 3: Individua l adjustments are positiv ely related to CSC. Past research has shown ASC to relate significantly to in-role performance (e.g., Becker et al., 1996). Interest ingly, in many cases, ASC has been found to relate more strongly to in-role performance than AOC (e.g., Becker & Kernan, 2003, Cheng et al., 2003; Vandenberghe et al., 2004). The proposed rationale for th is finding is that because supervisors have the formal authority to mon itor, direct, and provide feedback to their subordinates (Eisenberger, Stinglhamber, Va ndenberghe, Sucharski, & Rhoades, 2002) ASC may be especially salient in determ ining an employeeÂ’s in-role performance (Vandenberghe et al, 2004). However, it is lik ely that NSC relates to in-role performance in a similar direction (but to a lesser degree), because like affective commitment, normative commitment implies some level of concern for the target of commitment
10 (Meyer & Herscovitch, 2001). Existing rese arch also suggests that both ASC (e.g., Bentein, Stinglhamber, & Vandenberghe 2002) and NSC (Wasti & Can, 2008) are positively related to OCB directed toward the supervisor (OCB-supervisor). These relationships are not surprising, given that a past meta-analysi s revealed that non-specific OCB is significantly related to both AOC and NOC (Meyer et al., 2002). It intuits that ASC and NSC should relate to OCB-supervisor because if one is concerned with the well-being of their supervisor, one would be mo re inclined to engage in behaviors that benefit him or her. On the other hand, as discussed earlier, con tinuance commitment may not imply any behavior beyond simply remain ing associated with the target of the commitment. Thus, we would not expect that CSC relates to in-ro le performance or OCB-supervisor. Based on the discussion a bove, the following hypotheses are proposed: Hypothesis 4: ASC is positively related to a) in-role performance, and b) OCBsupervisor. Hypothesis 5: NSC is positively related to a) in-role performance, and b) OCBsupervisor. As stated earlier, staying with the orga nization is considered a focal outcome of organizational commitment. In line with this definition, all three bases of organizational commitment have been found to have negative relationships with tur nover (Meyer et al, 2002). However, it is likely that commitment to the supervisor also has favorable implications for maintaining organiza tional membership be cause discontinuing membership in the organization also involves the loss of the work relationship with the supervisor and coworkers. Thus, turnover may be a focal outcome of commitment to the organization, supervisor, and coworkers. Past research supports a negative relationship
11 between turnover intentions with ASC (e.g., Vandenberghe & Bentein, in press) and NSC (e.g., Stinglhamber et al., 2002), but has been equivocal for CSC (e .g., Stinglhamber et al., 2002). Nonetheless, the rationale of Meye r and Herscovitch (2001) suggests that all three bases of commitment s hould relate favorably to focal outcomes. Therefore, the following hypothesis is proposed: Hypothesis 6: Turnover intentions are ne gatively related to a) ASC, b) NSC, and c) CSC. Hypotheses 1-6 are summarized below in Figure 1.
12 Figure 1. Structural Model of Supervisor Commitment Note: Correlations between exogenous variab les are assumed. LMX = leader-member exchange; S-expect = supervisor expectati ons about staying with the organization; Adjust = individual adjustments to social positions; ASC = affective supervisor commitment; NSC = normative supervisor commitment; CSC = continuance supervisor commitment, Inrole perf = in-role performance; OCB-super = organizational citizenship behaviors directed toward the supervisor; Turn Intent = turnover intentions. A Structural Model of Coworker Commitment Although a sizable amount of research has examined commitment to the workgroup (see Riketta & Van Dick, 2005 for a meta-analysis), not much research has explored commitment to coworkers. The spar se research that doe s exist suggests that employees distinguish between their commitment s to coworkers from other foci and that coworker commitment relates positively to OCB (Snape et al., 2006, Wasti & Can, LMX S-expect Adjust ASC NSC CSC OCB-supe r Turn Intent Inrole per f + + + + + + + +
13 2008). Findings for commitment to the workgrou p have paralleled those of supervisor commitment. Workgroup commitment relates more strongly than does organizational commitment to workgroup-related outcomes, such a workgroup satisfaction and workgroup extra-role behaviors (Riketta & Van Dick, 2005). Furthermore, workgroup commitment can arise from perceptions of workgroup support (Bishop, Scott, Goldsby, & Cropazano, 2005). Although little research has been conducted on affective, normative, and continuance coworker commitment (or ACC, NCC, and CCC, respectively), it is possible that findings on organizational and wor kgroup commitment can be generalized to coworker commitment as well. Based on this assumption, hypotheses concerning coworker commitment parallel those put forth concerning supervisor commitment. Because it has been argued that affective commitment arises from a social exchange where the organization or workgroup gives th eir support in exchange for an employeeÂ’s AOC or workgroup commitment, respectively (Rhoades & Eisenberger, 2002; Bishop et al., 2005), it follows that a similar process ma y underlie coworker exchanges. Because this process involves some le vel of moral obligation (Was ti & Can, 2008), it is expected that both ACC and NCC would relate to pe rceived coworker support. Since normative commitment can result from socialization experiences which emphasize the importance of living up to othersÂ’ expect ations (Powell & Meyer, 2004) NCC may be influenced by perceived expectations from oneÂ’s cowork ers about staying with the organization. Finally, because continuance commitment arises from the perception of costs associated with leaving a position, it follows that CCC s hould be related to the perception that a job
14 change would require high amounts of individua l adjustment. Therefore, the following is proposed: Hypothesis 7: Perceived coworker support is positively related to a) ACC and b) NCC. Hypothesis 8: EmployeesÂ’ perceived exp ectations from coworkers about staying in the organization are positively related to NCC. Hypothesis 9: Individual adjustmen ts are positively related to CCC. Past research has supported that ACC and NCC relate significantly to OCB (Bryant, 2001; Wasti & Can, 2008). These findings, combined with general support for the compatibility hypothesis suggest that A CC and NCC will have positive effects on OCB directed toward coworkers (OCB-coworkers). Because of the motivational mindset which underlies continuance commitment, it is unlikely that CCC w ould be related to OCB-coworkers. Since turnover may be considered a focal outcome of commitment to coworkers, it is expected that all three ba ses of commitment will relate favorably to turnover intentions. Thus, the following hypotheses are presented. Hypothesis 10: OCB-coworkers is positi vely related to a) ACC and b) NCC. Hypothesis 11: Turnover intentions are negatively related to a) ACC, b) NCC, and c) CCC. Hypotheses 7-11 are summari zed below in Figure 2.
15 Figure 2. Structural Model of Coworker Commitment Note: Correlations between exogenous vari ables are assumed. PCS = perceived coworker support; Co-expect = coworker expectations about staying with the organization; Adjust = individual adjustme nts to social positions; ACC = affective coworker commitment; NCC = normative coworker commitment; CCC = continuance coworker commitment; OCB-co = organizational citizenship behaviors directed toward coworkers; Turn Intent = turnover intentions. Interactive Effects of Commitm ent to Interpersonal Foci The fact that employees experience vary ing levels of affective, normative, and continuance commitment simultaneously points to the need to consider whether the bases of commitment interact (Meyer & Hersc ovitch, 2001). When commitment to oneÂ’s organization is considered, numerous studies have found interactions (e.g., Gellatly et al., 2006; Jaros, 1997; Johnson et al., in pre ss; Meyer, Paunonen, Gellatly, Goffin, & Jackson, 1989; Randall, Fedor, & Longeneck er, 1990; Somers, 1995). As such, it is PCS Co-expect Adjust ACC NCC CCC OCB-co Turn Intent+ + + + + +
16 possible that interactive effects also characterize commitment to foci. Johnson, Groff, and Taing (in press) iden tified several models that potentially characterize interactions among commitment bases. A compensatory interaction model describes situations where a high level of onl y one commitment base is needed to bring about desirable work outcomes (see Figure 3). High levels of commitment for other bases are merely redundant. They posited that the compensatory model would hold when outcomes were focal attitudes or behaviors. This is because any base of commitment should be sufficient in itself to produce th e focal outcome (Meyer & Herscovitch, 2001). As stated earlier, maintaining membership in the organization may be seen as a focal behavior for commitment to supervisors and coworkers, because leaving the organization implies the loss of the work relationship with interpersonal foci within the organization. Figure 3. Compensatory Interaction Model Johnson et al. (in press) posited that the compensatory model does not predict outcomes that are discretionary (i.e. non-fo cal) to a commitment. Instead, the authors
17 proposed a synergistic model for such outcomes (see Figure 4). According to this model, the bases of commitment have non-redundant multiplicative effects on work outcomes such that the joint effects of high levels on multiple commitments have more favorable effects than is attainable by any one comm itment. This is because high levels of commitment for more than one base imply multiple reasons for performing a discretionary behavior. Figure 4. Synergistic Interaction Model Consistent with the reasoning of Johnson et al. (in press), Gellatly, Meyer, and Luchak (2006) found that for organizationa l commitment, the relationship between any base of commitment and staying intentions (a focal outcome) was strongest when the other bases of commitment we re low. However, for predicting OCB (a discretionary outcome), they found that those with high leve ls of all three bases were predicted to perform the most OCB. Based on the disc ussion concerning the compensatory and synergistic model in predicting focal and discretionary behaviors, the following
18 hypotheses are proposed: Hypothesis 12: When predicting turnover intentions, interact ions among the bases of supervisor commitment (i.e. ASC, NSC, and CSC) will show a compensatory pattern (i.e. high levels on more than one base of commitment are redundant in reducing turnover intentions) Hypothesis 13: When predicting (a) in-r ole performance and (b) OCB directed toward the supervisor, interactions am ong the different bases of supervisor commitment will show a syne rgistic pattern (i.e., outcomes are most favorable when employees report high levels on multiple bases). Hypothesis 14: When predicting turnover intentions, interact ions among the bases of coworker commitment (i.e. ACC, NCC, and CCC) will show a compensatory pattern (i.e. high levels on more than one base of commitment are redundant in reducing turnover intentions) Hypothesis 15: When predicting OCB direct ed towards coworkers, interactions among the different bases of coworker commitment (ACC, NCC, and CCC) will show a synergistic pattern. Interactive Effects of Comm itment across Different Foci Becker and Billings (1993) found that bei ng committed to more than one focus is beneficial. That is, being committed to multiple foci predicted the highest levels of satisfaction and prosocial beha vior. However, it is important to note that Becker and Billings (1993) did not actually examine intera ctive effects, but instead explored additive ones. To the authorÂ’s knowledge, the first test of statistical interactions between commitments to interpersonal foci was conduc ted by Snape et al. (2006) who posited that commitment to one focus is all that is needed to influence behavior. Therefore, commitments to additional foci would be larg ely redundant (i.e., a compensatory effect). They found some support for this, finding a compensatory interaction between ASC and affective work group commitment for two OC B dimensions (interpersonal harmony and protecting company resources). Most recently, Vandenberghe and Bentein (in press)
19 examined interactions between AOC and AS C for predicting turnover variables. They invoked LewinÂ’s (1943) field th eory, stating that ASC would be more salient to employees when AOC was low. They reasoned that high ASC in the presence of low AOC should exert stronger effects in reduc ing turnover because such context makes attachment to the supervisor more salient. In support of this, they found compensatory interactions for predic ting turnover in one sample and tu rnover intentions in two other samples. The results of Snape et al. (2006) and Vandenberghe and Bentein (in press) are somewhat consistent with the model outlin ed by Johnson et al. (in press). That is, Vandenberghe and Bentein (in press) found consistent evidence for a compensatory interaction for focal outcomes (turnover and turnover intent ions). Snape et al. (2006) found compensatory interactions for discretio nary outcomes (OCB) as well, but only two out of 30 interactions they test ed were significant. As such, more research is needed to determine the nature of interactions across foci. It may be the case that the compensatory model applies to focal outcomes, but interacti ons are absent for discretionary outcomes. The current study adopts the reasoning of Johnson et al. (in press) that the direction of interactions acro ss foci is determined by whethe r a behavior is considered focal or discretionary. However, as Becker and Billings (1993) noted, commitment to a specific focus should have positive implications concerning behavior and attitudes toward that focus, but not necessarily for othe r foci. Therefore, interactions between commitments to foci are more likely for outco mes that can be clearly related to each focus. As an example, supervisor commitment and coworker commitment may have
20 combined effects in determining OCB direct ed toward individuals (OCBI), since OCBI involves behavior toward both foci of commitment. Since OCBI is most likely viewed as a discretionary behavior, Johnson et al. (in press) w ould predict that the combined effects are synergistic. Similar rationale can be applied to predicting turnover intentions. Discontinuing membership in the organization is both related and focal to ASC, ACC, and AOC, because quitting the job also involves the loss of the work relationship with the supervisor and coworkers. Therefore, the co mbined effects of commitments to foci are likely to be compensatory when predicting turnover intentions. Although it is important not to ignore the bases of commitment when considering interactions between foci, only hypotheses pert aining to affective forms of commitment are proposed. This choice was influenced by a couple reasons. First, research on organizational commitment suggests that the a ffective base exerts the strongest effects on outcomes (Meyer et al., 2002). Second, consider ing all possible intera ctions between the three bases and three foci of commitment is simply not feasible. Examining three bases and three foci simultaneously suggests the possi bility of a 9-way interaction. I therefore limited my focus to affective commitment. Based on the discussion above, the following hypotheses are proposed: Hypothesis 16: The combined effects of ASC and ACC will be synergistic, when predicting OCBI. Hypothesis 17: The combined effect s of AOC, ASC, and ACC will be compensatory when predic ting turnover intentions. In summary, the current study investigat es affective, normative, and continuance commitment to supervisors and coworkers. Doing so is important because each respective
21 base may have unique antecedents and differe nt effects on outcomes. Furthermore, past research has revealed interactions among the bases of organizational commitment, raising the possibility that interactions also char acterize commitment to foci. If so, only measuring the affective base of commitment can result in model misspecification, thereby promoting inaccurate conclusions. Finally, si nce employees can feel attachment to multiple constituencies simultaneously, intera ctions across foci are also explored.
22 Chapter Two Method Participants and Procedure 241 employees working at least 20 h ours a week were recruited from undergraduate psychology courses to partic ipate in the current study. ParticipantsÂ’ average age was 22.40 ( SD = 5.38). They had been in college for an average of 3.47 years ( SD = 1.98) and had been employed at their current position for an average of 23.31 ( SD = 21.38) months. They worked an average of 28.69 hours per week ( SD = 8.72) and they were employed predominantly in retail/s ervice (e.g., cashier; 53.1 %) and professional industries (e.g., accounting; 15.1%). 2.9% repor ted working in a government agency (e.g., city hall), 1.3% reported a technical industry (e.g., mech anic), while 27.2% reported their sector as Â“otherÂ”. The majority of the sample were female (76.5%) and either Caucasian (60.3%), African American (17.2%), or Hispanic (11.7%). Participants received extra credit in th eir courses for completing the survey. In addition, they were asked to pass on a short surv ey to their supervisor and a coworker to complete. A cover letter was included with each other-report survey which stated that responses would be anonymous and to retu rn the survey using the self-addressed, stamped envelope that was provided to them. In order to discourage participants from completing the other-source surveys themselves respondents were told that they would
23 only receive extra credit for completing the self-report survey. Completion of the otherreport surveys did not lead to additional points. Furthermor e, contact information (e.g., phone number, email address) for supervisors wa s collected from the participant surveys. A subset of the supervisors (approximately 10%) who returned completed surveys was contacted to verify that they did indeed comp lete the supervisor survey. In all cases, the supervisors confirmed that they completed the survey. SupervisorsÂ’ average age was 37.81 ( SD = 11.62) and they worked an average of 44.62 hours per week ( SD = 9.68). 51.3% were male. They reported knowing thei r subordinate an average of 31.46 months ( SD = 54.23). The average age of coworkers was 28.64 ( SD = 11.10). 65.3% were female and they worked an average of 34.09 hours per week ( SD = 11.31). They reported knowing their coworker an average of 23.49 months ( SD = 36.50). The response rate for supervisors was 51.5%, while it was 49.4% for coworkers. Measures Except for the perceived expectations and individual adjustments scales, all survey items were measured via a 5-point Like rt scale (from 1 = Â“Str ongly disagreeÂ” to 5 = Â“Strongly agreeÂ”). Supervisors were asked to rate participantsÂ’ (their subordinate) inrole performance and OCB directed toward th e supervisor. Coworkers were asked to rate the participantsÂ’ (their coworker) OCB direct ed toward coworkers. All other measures were obtained from the participant. Organizational commitment Commitment to the organization was measured using Meyer and AllenÂ’s (1997) organizati onal commitment scale. Six items each measure AOC ( = .84), NOC ( = .86), and COC ( = .80). Sample items for AOC,
24 NOC, and COC, respectively, are Â“My organiza tion has a great deal of personal meaning for me,Â” Â“This organization deserves my l oyalty,Â” and Â“Right now staying with my organization is a matter of nece ssity as much as desire.Â” Supervisor commitment Commitment to the supe rvisor was assessed with Stinglhamber et al.Â’s (2002) supervisor commitment scale. The scale includes six items for ASC ( = .88), four items for NSC ( = .92), and five items CSC ( = .81). Sample items for ASC, NSC, and CSC, respectivel y, are Â“I feel proud to work with my supervisor,Â” Â“I would feel gui lty if I left my supervisor now,Â” and Â“Changing supervisors would necessitate that I acquire new work habits.Â” Coworker commitment Commitment to coworkers was measured with a modified version of Stinglhamber et al.Â’s (2002) workgroup commitment scale. Items were reworded by replacing instances of the wo rd Â“workgroupÂ” with Â“coworkers.Â” Six items each assessed ACC ( = .92), NCC ( = .93), and CCC ( = .90). Sample items for ACC, NCC, and CCC, respectively, are Â“I do not f eel emotionally attached to my coworkers ,Â” Â“I do not feel it would be right to leave my coworkers, even if it were to my advantage,Â” and Â“Changing coworkers w ould require a great deal of effort on my part to adapt to a new way of working.Â” Turnover intentions Employee intentions to leav e the organization were assessed using a hybrid scale ( = .88) consisting of three items developed by Mowday, Koberg, and McArthur (1984) and three items by M obley, Horner, and Hollingsworth (1978). A sample item is Â“I will probably l ook for a job in the near future.Â” LMX. Participants reported on their per ceptions of LMX quality, using Bernerth,
25 Armenakis, Feild, Giles, and Walk erÂ’s (2007) 8-item LMSX scale ( = .95). An example item is Â“When I give effort at work, my manager will return it.Â” Perceived coworker support Perceived coworker support ( = .86) was measured with a modified version of Eisenberger, Huntington, Hutchison, and SowaÂ’s (1986) 9item perceived organizational support scale. Items were reworded by replacing instances of the word Â“organizationÂ” with Â“coworkers.Â” An example item is Â“Even if I did the best job possible, my coworkers would fail to notice .Â” Perceived expectations about staying. Perceived expectations about staying in the organization were measured with an adap ted version of Powell and MeyerÂ’s (2004) expectation of others scale. Three items each were used to measure expectations from the supervisor ( = .86) and from coworkers ( = .85). Participants we re asked to rate each item in terms of their responsib ility for staying with the orga nization on a 5-point scale (1 = not at all responsible, 5 = very respons ible). Example items for supervisor and coworker expectations, respectively, are Â“Expect ations that my supervisor has for me to stayÂ” and Â“The need to return favors that my coworkers have done for me.Â” Individual adjustments. The perception that changi ng positions would involve individual adjustments was measured w ith Powell and MeyerÂ’s (2004) four-item individual adjustments to social positions scale ( = .82). Like the perceived expectations scales, participants were asked to respond to each item in terms of how responsible they were for the participant staying with the or ganization on a 5-point sc ale (1 = not at all responsible, 5 = very responsib le). An example item is Â“Time spent learning how to get along with people in the organizationÂ”.
26 In-Role Performance The participantÂ’s supervisor rated the subordi nateÂ’s in-role performance, using Williams and AndersonÂ’s (1991) seven-item in-role performance scale ( = .77). An example item is Â“Ade quately completes assigned duties.Â” OCB-supervisor. The participantÂ’s supervisor rated the subordinateÂ’s OCB directed toward the supervisor (OCB-supervisor; = .79). OCB-supervisor was be measured with two items taken from Bent ein, Stinglhamber, and Vandenberghe (2002), two items taken from Bryant (2001), and one item from Williams and AndersonÂ’s (1991) OCBI scale. An example item is Â“Informs me when an unforeseeable problem occurs on the job.Â” OCB-coworkers Coworkers rated participantÂ’s OCB directed toward coworkers (OCB-coworkers; = .89). OCB-coworkers measured with two items from Williams and AndersonÂ’s (1991) OCBI scale and three item s adapted from Podsakoff, Ahearne, and MackenzieÂ’s (1997) Helping behavior scale. The items from Podsakoff et al.Â’s (1997) scale address behavior toward crew members a nd were thus rewritten to reflect behavior toward coworkers. An example item from th is measure is Â“Is willing to share his/her expertise with other coworkers.Â” OCBI. OCB directed toward individuals (OCBI; = .86) was calculated using items from both the OCB-supervisor and OCB-coworkers scales. This approach was taken in order to capitalize on receiving da ta from both supervisors and coworkers. A similar approach has been adopted by others (e.g., Becker, 1992). Fu rthermore, OCBI is typically measured with items that tap O CB directed toward either coworkers or supervisors (e.g., Williams & Anderson, 1991).
27 Chapter Three Results Means, standard deviations, and corr elations among study va riables are reported in Table 1. As can be seen, the various bases and foci show substantial positive correlations. As would be expected, the hi ghest correlations tend to be among similar bases across foci (e.g., AOC and ACC), betw een affective and normative commitment to the same focus, and between normative and continuance commitment to the same focus. Although large, the correlations are not so high as to suggest complete conceptual overlap among bases or foci. Indeed, the observe d correlation between ACC and perceived coworker support and between ASC and LMX are higher than any of the intercorrelations among commitments. Before proceeding with tests of hypothes es, a multivariate analysis of variance (MANOVA) was conducted to examine equivalen ce in mean levels of responses between participants with only self-re port data and those from which other-report data was received. To do so, four groups were created: self-report only ( N = 102), supervisor and self-report ( N = 16), coworker and self-report ( N =28), and all three reports ( N = 98). The MANOVA compared the four groups on their reported levels of commitment, LMX, perceived coworker support, turnover inten tions, individual adjust ments, expectations from supervisors, and expectations from coworkers. The overall test revealed no significant differences between the groups, F (3, 45) = 1.37, ns
28 Table 1. Descriptive Statistics and Co rrelations among Study Variables VariableMSD12345678910111213141516171819 1. AOC 3.06.92(.84) 2. NOC 2.86.92 .66***(.86) 3. COC 3.03.87.11 .20**(.80) 4. ACC 3.82.90 .57***.35***.13*(.92) 5. NCC 2.381.06 .39***.56***.16*.39***(.93) 6. CCC 2.921.06.11 .15*.21***.30***.35***(.90) 7. PCS 3.83.66 .51***.31***-.06 .72***.29***.20**(.86) 8. Co_expect 3.101.11 .30***.35***.12 .32***.51***.15*.36***(.85) 9. ASC 4.03.83 .45***.35***-.01 .29***.13*-.03 .37***.21***(.88) 10. NSC 2.831.21 .36***.59***.11.12 .50***.03 .15*.32***.45***(.92) 11. CSC 2.86.98 .13*.18**.14*.14*.28***.46***.06.08 .19**.31***(.81) 12. LMX 3.70.93 .40***.36***.07 .25***.16*.03 .35***.21***.73***.47***.22***(.95) 13. S_expect 3.551.04 .38***.52***.13*.16*.32***-.04 .18***.52***.46***.55***.11 .51***(.86) 14. Adjust 3.421.01 .36***.39***.14*.25***.33***.14*.21***.42***.21***.27***.10 .26***.45***(.82) 15. Turn_Intent 2.811.05 -.54***-.50***-.12 -.26***-.16*-.05 -.25**-.11 -.42***-.32***-.08 -.41***-.29***-.22***(.88) 16. Inrole_perf 4.56.50 .19*.11.07.14-.13-.05 .21*.00 .35***.16.01 .33***.10.15-.13(.77) 17. OCB_super 4.46.60 .23*.10-.05 .21*-.01-.07 .27***.15 .41***.15.02 .32***.15.13-.15 .67***(.79) 18. OCB_co 4.48.62.08.03-.10.01-.09-.05.11.09 .22*.06-.08.17.17.10-.02 .33***.47***(.89) 19. OCBI 126.96.36.199.03-.11.05-.10-.09 .17*.12 .35***.06-.05 .26**.14.11-.04 .62***.84***.87***(.86) Note: For variables 1-15, N = 241. For variables 16-19, N ranges from 98 to 125. AOC = affectiv e organizational commitment; NOC = normative organizational commitment; COC = continuance organizational commitment; ACC = affective coworker commitment; NCC = normative coworker commitment; CCC = continuance coworker commitment; PCS = perceived coworker support; Co_expect = coworker ex pectations; ASC = affective supervisor commitment; NSC = normative supervisor commitment; CSC = continuance superv isor commitment; LMX = leader-member exchange; S_expect = supervisor expectations; Adjust = indivi dual adjustments; Turn_Intent = turnover inte ntions; Inrole_perf = in-role performance; OCB_super, OCB_co, and OCBI = organizationa l citizenship directed at supervisors, co workers, and individu als, respectively. *p < .05 **p < .01 *** p < .001
29 To determine whether employees actual ly distinguished between the various bases and foci of commitment, a confir matory factor analysis using AMOS 16.0 (Arbuckle, 2006) was conducted on all commit ment items (i.e., AOC, NOC, COC, ASC, NSC, CSC, ACC, NCC, and CCC items). The hypothesized nine-factor model was compared with several other models. These included a one-factor model, where all commitment items loaded on one factor, and two three-factor models. The three-factor foci model specified all supervisor commitmen t items as loading on a general supervisor commitment factor, all coworker commitmen t items as loading on a general coworker commitment factor, and all organizational commitment items as loading on a general organizational commitment factor. The threefactor bases model specified all of the affective commitment items as loading on a general affective commitment factor, all normative commitment items as loading on a general normative commitment factor, and all continuance commitment items as load ing on a general co ntinuance commitment factor. Finally, the nine-factor model specified items as loading on the factors they were designed to assess (e.g., AOC items loading ont o an AOC factor). A depiction of these models is shown in figure 5. As can be seen in Table 2, the hypot hesized model resulted in significantly better fit than the best-fitting three-factor model, 2 (33) = 2736.07, p < .001. Overall, the fit indices suggested passa ble fit for the hypothe sized model (RMSEA = .05, TLI = .84, CFI = .86). While the fit is no t perfect, it suggests that a nine-factor model is a substantial improvement over models that only distinguish between foci or only differentiate between bases.
30 Figure 5. Nested Confirmatory Factor Analytic Models Note: Correlations between factors are a ssumed. ASC = affective supervisor commitment; NSC = normative supervisor commitment; CSC = continuance supervisor commitment; AOC = affective organizational commitment; NOC = normative organizational commitment; COC = continuance organizational commitment; ACC = affective coworker commitment; NCC = normative coworker commitment; CCC = continuance coworker commitment
31 Table 2. Results of Confirmatory Factor A nalysis of Commitment Variables. Model 2df Change in 2TLICFIRMSEA Independence Model9686.371326 1-Factor6862.6712252823.70***.27.33.12 3-Factor Foci5649.6012211213.07***.43.47.10 3-Factor Bases5113.1112211749.56***.49.53.10 9-Factor Foci and Bases2377.0411882736.07***.84.86.05 Note: N = 241, TLI = Tucker-Lewis index; CFI = comparative fit index; RMSEA = root mean square error of approximation. Change in 2 for 1-factor model is relative to independence model, change in 2 for 3-factor models are relative to 1-factor model, and change in 2 for 9-factor model is relative to 3-factor bases model. ***p < .001 Structural Model of Supervisor Commitment Path analysis using maximum likelih ood estimation was conducted with AMOS 16.0 (Arbuckle, 2006) to test Hypotheses 16 on the sample of 112 matched employeesupervisor dyads. The initial model specified in Figure 1 had poor fit: 2 (22) = 79.50, RMSEA =.15, TLI =.61, CFI = .76. Modificat ion indices suggested adding a path between OCB-supervisor and in-role performa nce. A path leading from OCB-supervisor to in-role performance was favored rather than a path in the opposite direction for several reasons. First off, lab studies suggest that OCB influences ratings of overall performance (e.g., Allen & Rush, 1998; Rotundo & Sackett, 2 002). Second, it is plausible that helping supervisors may aid in clarifyi ng a subordinateÂ’s job role, whic h could in turn result in higher levels of in-role performance. Adding a path leading from OCB-supervisor to in-role performance resulted in significantly better fit to the data: 2 (1) = 54.27, p < .001. Overall, the model fit the data well: 2 (21) = 25.37, RMSEA =.04, TLI = .97, CFI = .98. The resulting standardized path coefficients for the model are shown in Figure 3, while results for significance tests of the
32 paths are shown in Table 3. Hypothesis 1 stated that LMX would be positively related to a) ASC and b) NSC and was fully suppor ted. Hypothesis 2, which stated that expectations from the supervisor about stay ing with the organization would be positively related to NSC, was supported because the path leading from supervis or expectations to NSC was significant and positive. Hypothesis 3, which stated that high perceptions of individual adjustments would show a positiv e relationship with CSC, was not supported. Hypothesis 4 predicted that AS C would be positively related to a) in-role performance and b) OCB-supervisor. This prediction received partial support as ASC was found to have a significant path leading to OCBsupervisor, but not in-role performance. Hypothesis 5 stated that NSC would also exert significant di rect effects on a) in-role performance and b) OCB-supervisor. This hypothesis was not supported because neither direct effect was signi ficant. Finally, hypothesi s 6 stated that turnover intentions would be negatively related to all three bases of supervisor commitment. This prediction was partially supported, in that ASC and NSC bot h had significant negative paths leading to turnover intentions, but CSC did not.
33 Figure 6. Standardized Estimates for Supe rvisor Structural Model Note: N = 112. Standardized regression coefficien ts are reported in the figure. LMX = leader-member exchange; S-expect = superv isor expectations about staying with the organization; Adjust = individual adjustments to social positions; ASC = affective supervisor commitment; NSC = normative supervisor commitment; CSC = continuance supervisor commitment, Inro le perf = in-role performance; OCBsuper = organizational citizen ship behaviors directed to ward the supervisor; Turn Intent = turnove r intentions. *p < .05 *** p < .001 LMX S-expect Adjust ASC NSC CSC OCB-supe r Turn Intent Inrole per f 0.03 .18* 0.06 -0.35*** .59*** 0.07 47* * 0.07 -.29*** .64*** .39*** 0.01 .38*** .42*** .21*
34 Table 3. Results for Supervisor Model Path Coefficients PathRaw Regression WeightStandard ErrorStandardized Regression Weight LMX --> ASC.52.07.60*** LMX --> NSC.25.12.18* S-expect --> NSC.55.10.47*** Adjust --> CSC.08.10.08 ASC --> Inrole perf.05.05.07 ASC --> OCB-super.31.07.39*** ASC --> Turn Intent-.47.11-.35*** NSC --> Inrole perf.02.03.06 NSC --> OCB-super.01.05.01 NSC --> Turn Intent-.25.07-.29*** CSC --> Turn Intent.03.08.03 OCB_super --> Inrole perf.52.06.63*** Note: N = 112. LMX = leader-member exchange; Sexpect = supervisor expectations about staying with the orga nization; Adjust = individua l adjustments to social positions; ASC = affective supervisor commitment; NSC = normative supervisor commitment; CSC = continuance supervisor commitment, Inrole perf = in-role performance; OCB-super = organizational citizenship behaviors directed toward the supervisor; Turn Intent = turnover intentions. *p < .05 **p < .01 *** p < .001 Structural Model of Coworker Commitment Path analysis was used to test H ypotheses 7-11 on the sample of 119 matched employee-coworker dyads. The initial hypothe sized model (See Figure 2) provided poor fit to the data: 2 (16) = 44.16, RMSEA = .12, TLI = .77, CFI = .87. Modification indices suggested adding a path leading from NCC to ACC. This is in line with Cohen (2007), who argued that normative commitment deve lops prior to entering the organization. Through early socialization experiences, peopl e develop moral values related to the importance of displaying loyalty to work or ganizations. These perceptions may then be further shaped by employeesÂ’ work-related e xperiences after they join a company. As such, it is plausible for NCC to develop befo re ACC and thus exert effects on it. While the addition of the path leading fr om NCC to ACC improved model fit 2 (1) = 12.56, p
35 < .001, overall model fit remained questionable: 2 (15) = 31.60, RMSEA = .10, TLI = .86, CFI = .92. Modification indices were agai n examined, which suggested the addition of a path between NCC and CCC. A path leading from NCC to CCC was favored over the reverse direction because it is conceivabl e that a particular cost associated with leaving coworkers may be feelings of guilt and shame. Additionally, as argued by Cohen (2007), normative commitment may devel op before affective or continuance commitment. The revised model resulted in si gnificantly better fit than the model which added a path leading from ACC to NCC: 2 (1) = 15.43 p < .001. Fit indices suggested that the revised model had good fit: 2 (14) = 16.17, RMSEA = .04, TLI = .98, CFI = .99. The resulting standardized path coefficients for the model are shown in Figure 6, while tests for the significance of the paths are shown in Table 4. Hypothesis 7 stated that perceived cowo rker support would be positively related to a) ACC and b) NCC. The path leading from coworker support to ACC was positive and significant. However, the path leadi ng from coworker support to NCC was not. Therefore, Hypothesis 7 received partial support. Hypothesis 8, which stated that coworker expectations would have a signi ficant path leading to NCC was supported. Hypothesis 9 was not supported because the path leading from individual adjustments to CCC was not significant. Hypothesis 10 was not supported because th e paths from ACC and NCC to OCB-coworkers were not significan t. Finally, Hypothesis 11 received partial support because ACC had a significant negative path leading to turnover intentions. However, neither NCC nor CCCÂ’s paths leadi ng to turnover intenti ons was significant.
36 Figure 7. Standardized Estimates for Co worker Commitment Model Note: N = 119. Standardized regression coefficien ts are reported in the figure. PCS = perceived coworker support; Co-expect = coworker expectations about staying; Adjust = individual adjustments to social positions; ACC = affective coworker commitment; NCC = normative coworker commitment; CCC = continuance coworker commitment; OCB-co = organizational citizenship behaviors directed toward coworkers; Turn In tent = turnover intentions. **p < .05 ** p < .001 Table 4. Results for Coworker Model Path Coefficients PathRaw Regression WeightStandard ErrorStandardized Regression Weight PCS --> ACC.88.08.67*** PCS --> NCC.04.12.03 Co-expect --> NCC.50.07.55*** Adjust --> CCC.10.09.10 ACC --> OCB-co.03.07.04 ACC --> Turn Intent-.35.10-.32*** NCC --> ACC.20.09.23** NCC --> OCB-co-.06.06-.10 NCC --> Turn Intent-.13.10-.13 NCC --> CCC.37.09.37*** CCC --> Turn Intent.13.10.13 Note: N = 119. PCS = perceived coworker support; Co-expect = coworker expectations about staying with the or ganization; Adjust = indivi dual adjustments to social positions; ACC = affective coworker commitment; NCC = normative coworker commitment; CCC = continuance coworker commitment; OCB-co = organizational citizenship behaviors dire cted toward coworkers; Turn Intent = turnover intentions. *p < .05 **p < .01 *** p < .001 PCS Co-expect Adjust ACC NCC CCC OCB-co Turn Intent0.13 -.14 .04 .03 .23*** .37*** .6 6 * .24* .35*** .47*** .10 .50*** -.32*** -.10
37 Interactive Effects among Bases of Commitment within a Single Focus For all hypotheses concerning interac tions, commitment variables were first centered before computing interaction te rms (Cohen, Cohen, West, & Aiken, 2003). After centering, hierarchical regression was used to test Hypotheses 12-15. In step 1, each commitment base was entered, while all possible 2-way interaction terms for each combination of bases were entered in step 2, followed by the 3-way interaction term in step 3. Demographic control variables were not used because a regression analysis revealed that gender, age, and tenure did not significantly predict any of the outcome variables in the current study. It should be noted that, initi ally, there was concern that multicollinearity would affect the ability to test interac tions. However, according to collinearity diagnostics, multicollinearity did not pose any problems (i.e. variance inflation factor numbers were much lower than 10 and tolerance numbers were all much greater than .10). Results for Hypotheses 12 and 13 are re ported in Table 5. Hypothesis 12 stated that interactions among bases of supervis or commitment would be compensatory for predicting turnover in tentions. This hypothesis was not supported, as none of the interaction terms were significant. Hypothesis 13 stated that interactions among the bases of supervisor commitment would be synergisti c for predicting a) in-role performance and b) OCB-supervisor. As can be seen in Ta ble 5, the only signifi cant interaction for predicting in-role performan ce was between ASC and CSC. However, not much can be made from this result for two reasons. As a set, the step 2 inte raction terms did not account for significant incremental variance in in-role performance. Additionally, a
38 follow-up regression model was performed wher e ASC and CSC were entered in step 1 and the ASC by CSC term was entered in step 2. Entered this way the ASC by CSC interaction failed to reach significance ( = -.17, ns ). As for OCB-supervisor, none of the interaction terms were significant. Ther efore, Hypothesis 13 was not supported. Table 5. Regression Results for Supervisor Commitment Base Interactions Predictor Step 1Step 2Step 3Step 1Step 2Step 3Step 1Step 2Step 3 ASC -.35***-.35***-.34***.33**.34**.32**.40***.40***.40*** NSC -.17*-.17*-.16*.06.04.02.02.02.03 CSC .04.04.05-.05.00-.03-.04-.01.00 ASC x NSC -.02-.03.06.08-.03-.03 ASC x CSC .01.00-.20*-.21*-.11-.11 NSC x CSC .03.04.07.04.02.03 ASC x NSC x CSC -.05.10-.03 Change in R2.00.00.03.01.01.00 Overall R2.20***.20***.20***.12**.16**.16**.17***.18**.18** Turnover IntentionsIn-role PerformanceOCB-supervisor Note: N = 241 for turnover intentions. N = 112 for in-role performance and OCBsupervisor. Standardized regression coeffi cients are reported in the table. ASC = affective supervisor commitment; NSC = normative supervisor commitment; CSC = continuance supervisor commitment *p < .05 **p < .01 *** p < .001 Results for Hypotheses 14 and 15 are pres ented in Table 6. Hypothesis 14 stated that interactions among bases of coworker commitment would be compensatory for predicting turnover in tentions. This hypothesis was not supported, as none of the interaction terms were significant. Hypothesis 15 stated that interactions among bases of coworker commitment would be synergisti c for predicting OCB-coworker. Again, none of the interaction terms were significant. Thus, Hypothesis 15 failed to receive support.
39 Table 6. Regression Results for Coworker Commitment Base Interactions Predictor Step 1Step 2Step 3Step 1Step 2Step 3 ACC -.24**-.26***-.24**.05.06.06 NCC -.09-.07-.06-.10-.14-.14 CCC .05.06.09-.03-.02-.08 ACC x NCC -.02-.01.02-.02 ACC x CCC .02.00-.06.00 NCC x CCC -.09-.08.11.09 ACC x NCC x CCC -.07.19 Change in R2.01.00.01.02 Overall R2.08***.08**.09**.01.02.04 Turnover IntentionsOCB-coworker Note: N = 241 for turnover intentions. N = 119 for OCB-coworker. Standardized regression coefficients are reported in the table. ASC = affective supervisor commitment; NSC = normative supervisor commitment; CSC = continuance supervisor commitment **p < .01 *** p < .001 Interactive Effects of Affective Commitment across Different Foci A similar approach was used to test Hypotheses 16 and 17. That is, commitment variables were centered before calculating interaction terms, and hierarchical regression was then used such that main effects were en tered in step 1 and in teraction effects were entered in step 2. Hypothesis 16 stated that the interaction between ASC and ACC would be synergistic for predicting OCBI. As can be seen in Table 7, the in teraction effect was not significant. Thus, Hypothe sis 16 was not supported. Hypo thesis 17 stated that the interaction between ASC, ACC, and AOC would be compensatory for predicting turnover intentions. Results from this analysis are also presented in Table 8. As can be seen, the 3-way interaction term was not sign ificant and therefore, Hypothesis 17 was not supported. Results of all hypotheses are summarized in Table 9.
40 Table 7. Regression Results for the Interac tion between Affective Supervisor and Coworker Commitment for Predicting Citizenship Behaviors Directed Toward Individuals PredictorStep 1Step 2 ASC.30**.31** ACC.10.10 ASC x ACC.01 Change in R2.00 Overall R2.13***.13** Note: N = 98. Standardized regression coefficien ts are reported in the table. ASC = affective supervisor commitment; ACC = affective coworker commitment. **p < .01 *** p < .001 Table 8. Regression Results for Tests of 3-Way Interactions Across Foci Predictor Step 1Step 2Step 3Step 1Step 2Step 3Step 1Step 2Step 3 ASC-.23***-.25***-.22**.32**.31**.41***.28*.29*.37** ACC.08.10.12.01-.03.11.06.02.11 AOC-.48***-.47***-.47***.05.04.04.09.10.09 ASC x ACC-.07-.11.12.02.11.05 ASC x AOC-.03-.03-.04-.04-.08-.09 ACC x AOC.06.06-.18-.24*-.07-.11 ASC x ACC x AOC-.09-.37**-.25 Change in R2.01.00.02.06**.01.02 Overall R2.33***.34***.34***.12**.14*.21**.13**.14*.16* Turnover IntentionsIn-role PerformanceOCB-Individuals Note: N = 241 for turnover intentions. N = 112 for in-role performance. N = 98 for OCB-individuals. Standardized regression co efficients are reported in the table. ASC = affective supervisor commitment; ACC = affective coworker commitment; AOC = affective organizational commitment *p < .05 **p < .01 *** p < .001
41 Table 9. Summary of Hypotheses and Results. HypothesisResult 1) LMX is positively related to a) ASC and b) NSCFully supported 2) EmployeesÂ’ perceived expectations from the supervisor about staying in the organization are positively related to NSC Fully supported 3) Individual adjustments are positively related to CSCNot supported 4) ASC is positively related to a) in -role performance, and b) OCB-supervisor Supported for OCB-supervisor 5) NSC is positively related to a) in-role pe rformance, and b) OCB-supervisorNot supported 6) Turnover intentions are negatively related to a) ASC, b) NSC, and c) CSC Supported for ASC and NSC 7) Perceived coworker support is positively related to a) ACC and b) NCC Supported for ACC 8) EmployeesÂ’ perceived expectations from coworkers about staying in the organization are positively related to NCC Fully supported 9) Individual adjustments are positively related to CCC.Not supported 10) OCB-coworkers is positively related to a) ACC and b) NCC.Not supported 11) Turnover intentions are negatively related to a) ACC, b) NCC, and CCC. Supported for ACC 12) When predicting turnover intentions, interactions among the bases of supervisor commitment (i.e. ASC, NSC, and CSC) will show a compensatory pattern (i.e. high levels on more than one base of commitment are redundant in reducing turnover intentions) Not supported 13) When predicting (a) in-role performan ce and (b) OCB dire cted toward the supervisor, interactions among the differen t bases of supervisor commitment will show a synergistic pattern (i.e., outcomes are mo st favorable when employees report high levels on multiple bases). Not supported 14) When predicting turnover intentions, interactions among the bases of coworker commitment (i.e. ACC, NCC, and CCC) will sh ow a compensatory pattern (i.e. high levels on more than one base of commitment are redundant in reducing turnover intentions) Not supported 15) When predicting OCB directed towards coworkers, interactions among the different bases of coworker commitment (ACC, NCC, and CCC) will show a synergistic pattern. Not supported 16) The combined effects of ASC and ACC will be synergistic, when predicting OCBI.Not supported 17) The combined effects of AOC, ASC, and ACC will be compensatory when predicting turnover intentions. Not supported Note: ASC = affective supervisor commitment; NSC = normative supervisor commitment; CSC = continuance supervis or commitment; LMX = leader-member exchange; ACC = affective coworker commitment; NCC = normative coworker commitment; CCC = continuance coworker commitment OCB-supervisor, OCBcoworkers, and OCBI = organizational ci tizenship directed at supervisors, coworkers, and indivi duals, respectively.
42 Exploratory Analyses Additional analyses were conducted to explore potential 3-way interactions concerning in-role performance and OCBI. Resu lts for these analyses are shown in Table 8. As can be seen, a significant 3-way inter action was found for in-role performance. The 3-way interaction term for OCBI was also si zable, but failed to reach significance. To better understand the interaction effect on in-role performance, mean values were plotted for individuals one standard deviation above and below the mean for commitment to each focus. Close examination of Figure 7 reveals that higher levels of ASC are associated with greater in-role pe rformance. However, the effect of ASC is more pronounced when individual shows high levels of either AOC or ACC, but not both. When individuals are high or low on both AOC and ACC, the pos itive effect of ASC is attenuated.
43 Figure 7. Three-Way Interaction between Commitments to Foci for Predicting In-role Performance -1.5 -1 -0.5 0 0.5 1 1.5 Low High ASCPredicted In-role Performance Low AOC, Low ACC Low AOC, High ACC High AOC, Low ACC High AOC, High ACC Note: AOC = affective organizational commitment, ASC = affective supervisor commitment, ACC = affective coworker commitment. High points show predicted standard scores for individuals one standard deviation above the mean on each respective commitment. Low points show predicted standard scores for individuals one standard deviation below the mean on each respective commitment.
44 Chapter Four Discussion The findings of the current study are threef old. First, they suggest that principles concerning bases of organizati onal commitment generalize to the bases of supervisor and coworker commitment. Second, unlike organi zational commitment, interaction effects among bases appear to be absent for commitme nt to interpersonal foci. Third, the current study adds evidence that interactions ex ist when multiple foci are concerned. Past research on commitment to differen t foci has focused almost exclusively on affective commitment. The current study a ddressed this shortcoming by examining normative and continuance commitment to supe rvisors and coworkers. Results revealed some generalizability of findings concerni ng the bases of organi zational commitment. First, as with organizational commitment, perceived support from a focus predicted higher levels of affective commitment to that focus. This was found to be true for both commitments to supervisors and coworker s. Additionally, theory on organizational commitment suggests that expectations a bout staying with the organization should influence normative commitment (Powell & Meyer, 2004). The current study confirmed this hypothesis when commitment to and exp ectations from supervisors and coworkers are concerned. Interestingly, th e perception that job change would require individual adjustments did not significantly predict con tinuance commitment in the path analysis. However, individual adjustments were related to CSC and CCC at the bivariate level.
45 The results of the current study also parallel findings on organizational commitment when the favorability of relationships between bases and outcomes are considered. Affective commitment to foci show ed the most favorable relationships with criteria. Normative commitment showed some positive effects, while there was no evidence of beneficial effects for continua nce commitment. When commitment to the supervisor is concerned, ASC significantly pr edicted higher levels of OCB-supervisor and lower turnover intentions, while NSC was only significantly related to lower turnover intentions. CSC was not signi ficantly related to either outcome. When coworker commitment is concerned, ACC predicted significantly lower levels of turnover intentions, while NCC and CCC did not. Research on organizational commitment sugge sts that statistical interactions exist among the bases of commitment but the cu rrent study suggests the absence of such interactions for supervisor and coworker co mmitment. However, it should be noted that the sample size available to test these interactions was not very large. Thus, the power to detect interactions may have been low. Furthe rmore, interactions can be difficult to detect using non-experimental methods (McClell and & Judd, 1993). On the other hand, the change in variance accounted for when interaction effects were added was never larger than .02, which suggests that if interactions exist they may not be practically important. More research will be needed before firm conclusions can be made about interactions among cases of commitment to foci. The current study also adds to a gr owing body of literature examining the combined effects of commitment to different foci. Evidence was found for a 3-way
46 interaction between ASC, ACC, and AOC for predicting in-role performance. However, the nature of this in teraction was counter to predictions made by Johnson et al. (in press). While in-role performance is certainly not a discretionary job requirement, it may not be necessarily be implied by commitment to vari ous foci. Therefore, Johnson et al. would predict the interaction to be s ynergistic. Instead, the interac tion was such that the highest levels of performance were predicted for cases when high ASC was combined with either high ACC or high AOC, but not both. High le vels of all three commitments actually predicted lower levels of in-role performance. Thus, the interaction was partly synergistic, but partly competitive. It should be noted that Johnson et al mention the possibility of competitive interactions where high levels of commitment to different foci actually work against each other, but they offer few predictions for wh en this would occur. They do suggest that such effect would be plausible when commitme nt to different foci force an employee to pursue incompatible goals. Pe rhaps an integration of th e reasoning put forth by Johnson et al. and the rationale stat ed by Vandenberghe and Bentein (in press) can explain the interaction. As stated by Johns on et al. it seems that commitment to multiple foci can be synergistic because employees have multiple reasons for performing a behavior. However, commitments to too many foci result in a reduction in salience of any particular focus (Vandenberghe & Bentein, in press), thereby producing a compensatory effect. Further, I propose that commitment to a large number of foci make competitive interactions increasingly likely because th ere will be a greater chance of opposing goals between foci. However, it seems that in c onsidering such inter actions the foci of
47 commitment are not interchangeable. In the ca se of in-role performance, the combined effect of ASC and either AOC or ACC was synergistic. But, an individual highly committed to all three foci was predicted to have much lower levels of performance, indicating a competitive effect. On the other hand, those with high levels of both AOC and ACC were predicted to have virtually the same level of performance as those highly committed to all three foci, indicating a comp ensatory effect. Whether the distinction between focal and discretionary behaviors predicts the nature of interactions across foci (vs. within foci) is yet to be seen. Implications and Directions for Future Research Although normative and continuance commitme nt to foci are distinguishable from affective commitment, the latter tended to s how the strongest relati onships with criteria. Furthermore, there was no evidence in the current study that the bases of commitment to foci interacted. This suggests that focu sing solely on affective commitment may be partially warranted because its effects are th e most robust of the three bases and they do not appear to depend on the relative levels of normative or continuance commitment. However, this is not to say that there is no value gained in consider ing the other bases of commitment. Indeed, NSC predicted lower le vels of turnover intentions when the contribution of ASC was accounted for. Th is suggests that at least normative commitment can aid in the incremental predic tion of criteria. Furthe rmore, it seems that each commitment base is predicted by different antecedents. Thus, if oneÂ’s interest is in understanding how to improve outcomes via co mmitment to foci, consideration of bases is useful (i.e., focus on antecedents which ar e likely to increase affective commitment).
48 The current study also sugge sts robust effects of ASC, relative to AOC and ACC. Indeed, the multiple regression analyses indi cate that ASC was positively related to inrole performance, OCBI, and turnover intentio ns even when the effects of AOC and ACC are accounted for. This suggests that the infl uence of the supervisor is central to the behavior and attitudes of employees. As st ated earlier, supervisors have the formal authority to monitor and direct employees (Eisenberger et al., 2002), which may make supervisors an especially salient target of commitment. Therefore, researchers and practitioners alike may want to focus on ways to foster ASC. The current study also adds evidence fo r the existence of interactions across commitments to foci. This makes prediction of the effects of commitment more complicated for both researchers and practitioners Therefore, future research is needed to determine for what foci and outcomes these in teractions exist. Furthermore, theory is needed to explain under what conditions a pa rticular pattern of interactions will be observed. As stated earlier, it is unclear wh ether explaining interac tions across foci using the distinction between discretionary and focal outcomes will hold. Similarly, the idea that multiple commitments automatically decrease the influence of any one commitment focus does not seem sufficient to explain such interactions. What seems clear is that synergistic, competitive, and compensato ry interactions are all possible. That coworker commitment did not si gnificantly predicted OCB-coworker was somewhat surprising, given a past meta-ana lysis revealed workgroup commitment to be significantly related to workgr oup extra-role behaviors (Rik etta & van Dick, 2005). This suggests that commitment to the workgroup is not necessarily the same as commitment to
49 coworkers. According to Thompson (2004), workgroups imply some level of interdependence and shared responsibili ty between members of the group. This interdependence is not implied when consid ering the more gene ral distinction of coworkers. Therefore, the effects of cowo rker commitment may be moderated by other factors, such as interdependence or group c ohesion. As interdependence increases, the effect of coworker commitment may b ecome more salient to employees. Finally, another interest ing feature of the result s was that continuance commitment did not significantly predict tur nover intentions despite theory stating that turnover is a focal outcome of employee co mmitment. Recent research on COC (e.g., Groff, Granger, Taing, Jackson, & Johns on, 2008; Vandenberghe et al., 2007) has suggested that it is comprised of two dis tinct dimensions: COC based on the lack of employment alternatives and COC based on th e perception that job change would involve high sacrifices. When separated as such, research suggests that COC based on low alternatives relates positively to turnover intentions while COC based on high sacrifices relates negatively (Stinglhamber et al., 2002) Therefore, combining the two dimensions into a unidimensional measure may cancel out each respective factorÂ’s relationship with turnover intentions. It remains an open quest ion whether continuance commitment to foci is also multidimensional. If so, a similar e xplanation may underlie th e lack of significant relationships between con tinuance commitment to foci and turnover intentions. Limitations A limitation of the current research is th e use of data from university students. This may call into question whether the resu lts reported here gene ralize to non-students,
50 which may systematically differ in age and te nure. Even so, it is not clear how age or tenure would affect the nature of the results. On the positive side, in contrast to data collected from non-students in a single organi zation, the use of a student sample allows for data concerning the effects of commitment for a diverse set of jobs and organizations, thereby increasing generalizability to some extent. Nonetheless, data from an older sample would mitigate concerns over genera lizability. Additionally, because the data reported here is cross-sectional, causal infere nces cannot be drawn. Ho wever, past studies examining the effects of commitment ha ve employed longitudi nal designs and found results which support a similar causal order of variables as reported here (e.g. Bentein et. al, 2002, Stinglhamber & Vandenberghe, 2003; Vandenberghe & Bentein, in press). One may argue that the observed results are explainable by particular statistical artifacts and biases. For example, some predictors and outcomes (e.g., commitment and turnover intentions) were reported by the sa me source and thus, results may have been influenced by common method bias. The resu lts for performance may be explained by mutual liking between the employee and the focus reporting on the employeeÂ’s performance. For example, an employee may sh ow attachment to the supervisor and thus the supervisor evaluates the employeeÂ’s performance favorably even if the employee does not truly perform at high levels. Resu lts concerning ASC provide an example for why such phenomena may not hold. Although not reported, I also conducted an analysis where OCB-coworker was regressed on ASC, AOC, and ACC. In this analysis, ASC, which involves attachment to the supervis or, was reported by the employee while OCB was reported by the coworker. Even so, ASC was significantly related to OCB-coworker,
51 when AOC and ACC were held constant. Inte restingly ACC did not significantly predict OCB-coworker. As such, the positive effect s of ASC cannot easily be explained away with biases. Conclusion Past research has demonstrated the importance of examining commitment to interpersonal foci within the organizati on, such as supervisors and coworkers. Unfortunately, that research had focused al most exclusively on affective bases of commitment. The current study addressed this shortcoming by examining all three bases of commitment to supervisors and coworkers. The results revealed that relationships between the bases and othe r criteria correspond well with findings on organizational commitment. Unlike organizational commitme nt, no evidence was found for interactions among bases for commitment to supervisors and coworkers. Finally, the current study adds to evidence that interac tions exist among commitments to foci. However, the nature of this interaction diverged from previous fi ndings. Thus, future research is needed to establish theory concerning inter actions among commitments to foci.
52 References Ajzen, I., & Fishbein, M. (1977) Attitude-behavior relations: A theoretical analysis and review of empirical research. Psychological Bulletin, 84, 888-918. Allen, N. J., & Meyer, J. P. (1990). The m easurement and antecedents of affective, continuance, and normative comm itment to the organization. Journal of Occupational Psychology, 63, 1-18. Allen, T. D., & Rush, M. C. (1998). The effect s of organizational citizenship behavior on performance judgments: A field stud y and a laboratory experiment. Journal of Applied Psychology, 83, 247-260. Arbuckle, J. L. (2006) Amos 7.0 userÂ’s guide Chicago, IL: SPSS Inc. Becker, T. E. (1992) Foci and bases of comm itment: Are they distinctions worth making? Academy of Manageme nt Journal, 35, 232-244. Becker, T. E., & Billings, R. S. (1993) Pr ofiles of commitment: An empirical test. Journal of Organizational Behavior, 14, 177-190. Becker, T. E., & Kernan, M. C. (2003). Matching commitment to supervisors and organizations to in-role a nd extra-role performance. Human Performance, 16, 327-348. Becker, T. E., Billings, R. S., Eveleth, D. M., & Gilbert, N. L. (1996). Foci and bases of employee commitment: Implications for job performance. Academy of Management Journal, 39, 464-482.
53 Bentein, K., Stinglhamber, F., Vandenberghe, C. (2002). Organization, supervisor-, and workgroup-directed commitments and citi zenship behaviours: A comparison of models. European Journal of Work and Organizational Psychology, 11, 341362. Bernerth, J. B., Armenakis, A. A., Feild, H. S., Giles, W. F., & Walker, H. J. (2007). Leader-member social exchange (LMSX): Development and valid ation of a scale Journal of Organizational Behavior, 28, 979-1003 Bishop, J. W., Scott, K. D., Goldsby, M. G., & Cropanzano, R. (2005). A construct validity study of commitment and per ceived support variables: A multifocal approach across different team environments. Group and Organization Management, 30, 153-180. Borman, W. C., & Motowidlo, S. J. (1993). Expanding the criterion domain to include elements of contextual performance. In N. Schmitt & W. Borman (Eds.), Personnel selection in organizations (pp. 71Â–98). New York: Jossey-Bass. Bryant, M. E. (2001). The nature of empl oyee commitment: Explori ng the value of the multidimensional perspective Dissertation Abstracts In ternational: Section B: The Sciences and Engineering, 62, 2980. Chen, Z. X., Tsui, A. S., & Farh, J. L. ( 2002). Loyalty to supervis or vs. organizational commitment: Relationships to employee performance in China. Journal of Occupational and Organizational Psychology, 75, 339-356. Cheng, B.-S., Jiang, D.-Y., & Riley, J. H. (2003). Organizational commitment, supervisory commitment, and employee outcomes in the Chinese context:
54 Proximal hypothesis or global hypothesis? Journal of Organizational Behavior, 24, 313-334. Cohen, A. (2007). Commitment before and afte r: An evaluation and reconceptualization of organizational commitment. Human Resource Manage ment Review, 17, 336354. Cohen, J., Cohen, P., West, S., & Aiken, L. (2003). Applied multiple regression/correlation analysis for th e behavioral sciences (3rd Ed.). Hillsdale, NJ: Lawrence Erlbaum Associates. Cohen-Charash, Y., & Spector, P. E. (2001). The role of justice in organizations: A metaanalysis. Organizational Behavior and Hu man Decision Processes, 86, 278-321. Colquitt, J. A., Conlon, D. E., Wesson, M. J., Porter, C. O. L. H., & Ng, K. Y. (2001). Justice at the millennium: A meta-analytic review of 25 years of organizational justice research. Journal of Applied Psychology, 86, 425-445. Deci, E. L., & Ryan, R. M. (1985). Intrinsic motivation and self-determination in human behavior New York: Plenum. Eisenberger, R., Huntington, R., Hutchi son, S., & Sowa, D. (1986). Perceived organizational support. Journal of Applied Psychology, 71, 500-507. Eisenberger, R., Stinglhamber, F., Vandenber ghe, C., Sucharski, I., & Rhoades, L. (2002). Perceived supervisor support: C ontributions to perceived organizational support and employee retention. Journal of Applied Psychology, 87, 565-573. Farh, J. L., Early, P. C., & Lin, S. C. (1997). Impetus for action: A cultural analysis of justice and organizational citizensh ip behavior in Chinese society. Administrative
55 Science Quarterly, 42, 421-444. Gagn, M., & Deci, E. L. (2005). Self-det ermination theory and work motivation. Journal of Organizational Behavior, 26, 331-362. Gellatly, I. R., Meyer, J. P., & Luchak, A. A. (2006). Combined effects of the three commitment components on focal and discre tionary behaviors: A test of Meyer and Herscovitch's propositions. Journal of Vocational Behavior, 69, 331-345. Groff, K., Granger, B., Taing, M., Ja ckson, E., & Johnson, R. E. (2008, August). Exploring continuance commitment: A multidimensional approach Paper presented at the 68th Academy of Management Annual Meeting, Anaheim, California. Grolnick, W. S., & Ryan, R. M. (1987). Autonomy in children's learning: An experimental and individual difference investigation. Journal of Personality and Social Psychology, 52, 890-898. Jaros, S. J. (1997). An assessment of Meye r and Allen's (1991) th ree-component model of organizational commitment and turnover intentions. Journal of Vocational Behavior, 51, 319-337. Johnson, R. E., Groff, K. W., & Taing, M. U. (i n press) Nature of the interactions among organizational commitments: Non-exis tent, competitive, or synergistic? British Journal of Management. Lewin, K. (1943). Defining the Â‘field at a given timeÂ’. Psychological Review, 50, 292310. Mathieu, J. E., & Zajac, D. M. (1990). A re view and meta-analysis of the antecedents,
56 correlates, and consequences of organizational commitment. Psychological Bulletin, 108 171-194. McClelland, G. H., & Judd, C. M. (1993). Statis tical difficulties of detecting interactions and moderator effects. Psychological Bulletin, 114, 376-390. Meyer, J. P., & Allen, N. J. (1984). Testi ng the side-bet theory or organizational commitment: Some methodological considerations. Journal of Applied Psychology, 69, 372-378. Meyer, J. P., & Allen, N. J. (1991). A three-component conceptualization of organizational commitment. Human Resource Management Review, 1, 61-89. Meyer, J. P., & Allen, N. J. (1997). Commitment in the workplace: Theory, research, and application Thousand Oaks: Sage. Meyer, J. P., Becker, T. E., & Vandenbe rghe, C. (2004) Employee commitment and motivation: A conceptual analysis and integrative model. Journal of Applied Psychology, 89, 991-1007. Meyer, J. P., & Herscovitch, L. (2001). Co mmitment in the workplace: Toward a general model. Human Resource Management Review, 11, 299-326. Meyer, J. P., Paunonen, S. V., Gellatly, I. R., Goffin, R. D., & Jackson, D. N. (1989). Organizational commitment and job performance: It's the nature of the commitment that counts. Journal of Applied Psychology, 74, 152-156. Meyer, J. P., Stanley, D.J., Herscovitc h, L., & Topolnytsky, L. (2002). Affective, continuance, and normative commitment to the organization: A meta-analysis of antecedents, correlates, and consequences. Journal of Vocational Behavior, 61
57 20-52. Mobley, W. H., Horner, S. O. & Hollings worth, A. T. (1978). An evaluation of precursors of hospital employee turnover. Journal of Applied Psychology, 63, 493-522. Mowday, R. T., Koberg, C. S. & McArthur, A. W. (1984). The psychology of withdrawal process: a cross-validational test of M obley's intermediate linkages model of turnover in two samples. Academy of Management Journal, 27, 79-94. Organ, D. W., & Ryan, K. (1995) A meta-analyti c review of attitudinal and dispositional predictors of organizationa l citizenship behavior. Personnel Psychology, 48, 775802. Podsakoff, P. M., Ahearne, M., & MacKenzie, S. B. (1997). Organizational citizenship behavior and the quantity and qual ity of work group performance. Journal of Applied Psychology, 82, 262-270. Powell, D. M. & Meyer, J. P. (2004). Side-bet theory and the three-component model of organizational commitment. Journal of Vocational Behavior 65 157-177. Randall, D. M., Fedor, D. B., & Longenecker, C. O. (1990). The behavioral expression of organizational commitment. Journal of Vocational Behavior, 36, 210Â–24. Reichers, A. E. (1985). A review and recon ceptualization of organizational commitment. Academy of Management Review, 10 (3), 465-476. Rhoades, L., & Eisenberger, R. (2002). Percei ved organizational support: A review of the literature. Journal of Applied Psychology, 87, 698Â–714. Riketta, M., & Van Dick, R. ( 2005). Foci of attachment in organizations: A meta-analytic
58 comparison of the strength and correlat es of workgroup versus organizational identification and commitment. Journal of Vocational Behavior, 67, 490-510. Rotundo, M., & Sackett, P. R. (2002). The re lative importance of ta sk, citizenship, and counterproductive performance to global ratings of job perf ormance: A policycapturing approach. Journal of Applied Psychology, 87, 66-80. Ryan, R. M., & Connell, J. P. (1989). Perceive d locus of causality and internalization: Examining reasons for acting in two domains. Journal of Personality and Social Psychology, 57, 749-761. Shore, L. M., Tetrick, L. E., Lynch, P., & Barksdale, K. (2006) Social and economic exchange: Construct development and validation. Journal of Applied Social Psychology, 36(4), 837-867. Snape, E., Chan, A. W., & Redman, T. (2006) Multiple commitments in the Chinese context: Testing compatibility, cu ltural, and moderating hypotheses. Journal of Vocational Behavior, 69, 302-314. Somers, M. J. (1995). Organizational comm itment, turnover and absenteeism: An examination of direct and interaction effects. Journal of Organizational Behavior, 16, 49-58. Stinglhamber, F., Bentein, K., & Vandenber ghe, C. (2002) Extension of the threecomponent model of commitment to five foci. European Journal of Psychological Assessment, 18, 123-138. Stinglhamber, F., & Vandenberghe, C. (2003) Organizations and supervisors as sources of support and targets of commitment: A longitudinal study. Journal of
59 Organizational Behavior, 24, 251-270. Thompson, L. L. (2004). Making the team. A guide for managers (2nd ed.). Upper Saddle River, NJ: Prentice Hall. Vandenberghe, C., & Bentein, K. (in press) A closer look at the relationship between affective commitment to supervisors and organizations and turnover. Journal of Occupational and Organizational Psychology. Vandenberghe, C., Bentein, K., & Stinglhamber F. (2004) Affective commitment to the organization, supervisor, and wor kgroup: Antecedents and outcomes. Journal of Vocational Behavior, 64, 47-71. Vandenberghe, C., Bentein, K., Michon, R., Cheb at, J., Tremblay, M., & Fils, J. (2007). An examination of the role of perc eived support and employee commitment in employee-customer encounters. Journal of Applied Psychology, 92(4), 11771187. Wasti, S. A., & Can, O. (2008) Affective and normative commitment to organization, supervisor and coworkers: Do collectivist values matter? Journal of Vocational Behavior, 73, 404-413. Williams, L. J., & Anderson, S. E. (1991) Job satisfaction and organizational commitment as predictors of organizati onal citizenship and in-role behaviors. Journal of Management, 17, 601-617.
61 Appendix A: List of Survey Items Affective Organiza tional Commitment 1. I would be happy to spend the rest of my career with my current organization 2. I really feel as if my or ganizationÂ’s problems are my own 3. I do not feel like Â‘part of the familyÂ’ at my organization 4. I do not feel Â‘emotionally a ttachedÂ’ to my organization 5. My organization has a great d eal of personal meaning for me 6. I feel a strong sense of belonging to my organization Normative Organizational Commitment 1. I do not feel any obligation to remain with my current employer 2. Even if it were to my advantage, I do not feel it would be right to leave my organization now. 3. I would feel guilty if I left my organization now 4. This organization deserves my loyalty 5. I would not leave my organization right now because I have a sense of obligation to the people in it 6. I owe a great deal to this organization
62 Appendix A: (Continued) Continuance Organizational Commitment 1. It would be very hard for me to leave my organization right now, even if I wanted to 2. Too much in my life would be disrupted if I decided I wanted to leave my organization now 3. Right now staying with my organization is a matter of necessity as much as desire 4. I feel that I have too few options to consider leaving my organization 5. One of the few serious consequences of l eaving my organization w ould be the scarcity of available alternatives 6. One of the major reasons I continue to work for my organization is that leaving would require considerable personal sacrificeÂ—anot her organization may not match the overall benefits that I have here Affective Supervisor Commitment 1. I have respect for my supervisor 2. I appreciate my supervisor 3. I have little admiration for my supervisor 4. I feel proud to work with my supervisor 5. My supervisor means a lot to me 6. I do not really feel attached to my supervisor
63 Appendix A: (Continued) Normative Supervisor Commitment 1. I would feel guilty if I left my supervisor now 2. I feel I have a moral obligation to continue working with my supervisor 3. I would not leave my supervisor at the mo ment because I feel obligated to him/her 4. If I were offered the chance to work w ith another supervisor I would not think it morally right to leave my current supervisor Continuance Supervisor Commitment 1. Changing supervisors would require me to substantially re-organize the way I perform my job 2. Changing supervisors would demand a great deal of effort on my part order for me to adapt to a new leadership style 3. Changing supervisors would necessitate that I acquire new work habits 4. I am so used to working with my current s upervisor that it would be difficult for me to change 5. There would be few modifications to th e way I work if I changed supervisors
64 Appendix A: (Continued) Affective Coworker Commitment 1. My coworkers mean a lot to me 2. I really feel a sense of belonging with my coworkers 3. I feel proud to be asso ciated with my coworkers 4. I do not feel a strong sense of belonging with my coworkers 5. I do not feel like part of the family with coworkers 6. I do not feel emotionally attached to my coworkers Normative Coworker Commitment 1. It would not be morally right for me to leave my coworkers right now 2. I do not feel it would be right to leave my coworkers now, even if it were to my advantage 3. I think I would be guilty if I left my coworkers now 4. I feel I have to continue to work with my coworkers 5. If I were offered another j ob with different coworkers, I would not feel it was right to leave them 6. I would not leave my coworkers right now because I feel obligated to certain ones
65 Appendix A: (Continued) Continuance Coworker Commitment 1. Changing coworkers would require me to adjust to new work habits 2. Changing coworkers would require me to get used to a new organization of work 3. If I changed coworkers, I would have to re-adapt to new group norms 4. Changing coworkers would require a great deal of effort on my part to adapt to a new way of working 5. Changing coworkers would require me to completely re-organize the way I work 6. I am so used to working with my present coworkers that it would be difficult for me to change Leader-Member Exchange 1. My supervisor and I have a two-way exchange relationship 2. I do not have to specify the exact conditions to know my supervisor will return a favor 3. If I do something for my supervisor, he or she will definitely repay me 4. I have a balance of inputs and outputs with my supervisor 5. My efforts are reciprocated by my supervisor 6. My relationship with my supervisor is co mposed of comparable exchanges of giving and taking 7. When I give effort at work, my supervisor will return it 8. Voluntary actions on my part will be returned in some way by my supervisor
66 Appendix A: (Continued) Perceived Coworker Support 1. My coworkers strongly cons iders my goals and values 2. Help is available from my coworkers when I have a problem 3. My coworkers really care about my well-being 4. My coworkers are willing to extend themselv es in order to help me perform my job to the best of my ability 5. Even if I did the best job possibl e, my coworkers would fail to notice 6. My coworkers care about my general satisfaction at work 7. My coworkers show very little concern for me. 8. My coworkers care about my opinions 9. My coworkers take pride in my accomplishments at work Individual Adjustments to Social Positions 1. Time spent learning the policies a nd procedures of the organization 2. Time spent learning how to get al ong with people in the organization 3. Training IÂ’ve received th at wouldnÂ’t be useful in another organization 4. Time spent learning how to adjust to the performance expectations at the organization Perceived Expectations from the Supervisor about Staying 1. Expectations that my supervisor has for me to stay 2. My supervisor counting on my con tinued employment at the organization 3. My obligation to recipr ocate things my supervisor has done for me
67 Appendix A: (Continued) Perceived Expectations fr om Coworkers about Staying 1. My coworkers counting on me to stay with the organization 2. Expectations that my coworkers have for me to stay 3. The need to return favors that my coworkers have done for me Turnover Intentions 1. I constantly think about quitting 2. All things considered, I woul d like to find a comparable jo b in a different organization 3. I will probably look for a new job in the near future 4. I will probably find an acceptable alternative if I look for a new job 5. I am unlikely to leave my job soon 6. I donÂ’t have any intention to look for a new job Organizational Citizenship Behaviors directed toward the Supervisor 1. Gives advance notice to me wh en unable to come to work 2. Informs me when an unforeseeable problem occurs on the job 3. Assists me with my work (when not asked) 4. Volunteers for tasks that will help me do my job 5. Does what I ask without complaining
68 Appendix A: (Continued) In-role Performance 1. Adequately completes assigned duties 2. Fulfills responsibilities speci fied in job description 3. Performs tasks that ar e expected of him/her 4. Meets formal performance requirements of the job 5. Engages in activities that will directly affect his/her perf ormance evaluation 6. Neglects aspects of the job he/s he is obligated to perform 7. Fails to perform essential duties Organizational Citizenship Behavio rs directed toward Coworkers 1. Willing to give their time to help co -workers with work-related problems 2. Willing to share their expertise with other co-workers 3. Helps co-workers out if someone is falling behind on their work 4. Takes time to listen to co -workersÂ’ problems and worries 5. Passes along information to co-workers