لخّصلي

Journal of Occupational and Organizational Psychology (2012), 85, 421-432
C 2011 The British Psychological Society
Short research note
Leader and follower organizational identification: The mediating role of leader behaviour and implications for follower OCB
Sebastian C. Schuh1*, Xin-an Zhang2, Nikolai W. Egold1, Matthias M. Graf1, Diwakar Pandey3 and Rolf van Dick1
1Goethe University, Frankfurt, Germany
2Shanghai Jiao Tong University, Shanghai, China
3Universal College of Medical Sciences, Bhairahawa, Nepal
The transfer model of organizational identification (OI) posits a trickle-down process of OI from leaders to followers.Despite this cumulated evidence, however, two important questions have remained unanswered: first, it is still unclear how leaders affect their employees' OI. Even though van Dick et al. (2007) suggested that transformational leadership (TFL) behaviours such as showing dedication to the cause of the group may play a central role in this process, none of the previous studies have examined the behavioural links that connect leader OI and follower OI. More specifically, while previous research has found a positive relationship between TFL and follower OI (e.g., Kark, Shamir, & Chen, 2003) as well as between follower OI and followers' group-oriented efforts (e.g., Riketta, 2005), this is, to our best knowledge, the first study that examines the link between leader OI and (group-oriented) leadership behaviour.See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

424 Sebastian C. Schuh et al.
(83 supervisors and 437 followers), Walumbwa, Avolio, and Zhu (2008) reported that employees' OI mediated the link between TFL and employee performance.Transformational leadership
L1: TFLij = 0j + 1j(F_ageij) + 2j(F_genderij) + 3j(F_ageijL_agej) + 4j(F_genderijL_genderj) + rij L2: 0j = 00 + 01(L_agej) + 02(L_genderj) + 03(L_OIj) + u0j
L2:1j =10 2j =20 3j =30 4j =40
Follower age ( 10 ) Leader age (01) Follower gender ( 20 ) Leader gender ( 02 ) F_age x L_age ( 30 ) F_gender x L_gender Leader OI ( 03 )
-.25 .19
-.08 .09
.19 .26
-.18 .19
.09 .05
-.07 .21
.25
-.01 -.01 -.79 -.62
.00 .62
.07 .02 .62 .34 .00 .50 .21
.04 .05 .45 .20 .00 .37 .13 .07
CI indirect effect (95%) ~R2
.40 .10
[.083, .085] .41
( 40 )
* **
~R2
2.Follower OI
L1: F_OIij = 0j + 1j(F_ageij) + 2j(F_genderij) + 3j(F_ageijL_agej) + 4j(F_genderijL_genderj)
.14
+ 5j(TFLij) + rij
L2: 0j = 00 + 01(L_agej) + 02(L_genderj) + 03(L_OIj) + u0j L2:1j =10 2j =20 3j =30 4j =40 5j =50
.24
Follower age ( 10 ) Leader age ( 01 ) Follower gender ( 20 ) Leader gender ( 02 ) F_age x L_age ( 30 ) F_gender x L_gender Leader OI ( 03 )
TFL (50)
CI indirect effect (95%)
.03 -.09 .15 .25 -.03 -.18 .05 .32
.22 .10 .31 .20 .06 .24 .11 .08
.05
.08 -.34 -.53 .00 .23 -.02 .28
( 40 )
( 40 )
**
**
[.016, .168] .11
[.075, .386] .28
~R2
3.Follower OCB
L1: F_OCBij = 0j + 1j(F_ageij) + 2j(F_genderij) + 3j(F_ageijL_agej) + 4j(F_genderijL_genderj)
+ 5j(TFLij) + 6j(F_OIij) + rij
L2: 0j = 00 + 01(L_agej) + 02(L_genderj) + 03(L_OIj) + u0j
L2:1j =10 2j =20 3j =30 4j =40 5j =50 6j =60 Follower age (10)
Leader age (01)
Follower gender (20)
Leader gender (02) F_age x L_age (30) F_gender x L_gender Leader OI ( 03 )
TFL (50) Follower OI ( 60 )
.00 .05
.01 .05
.06 .45
*
.00 .00
-.17 .37
* **
Note.[Corrections added on 14 December 2011 after initial online publication on 7 November 2011: some data were corrupted during typesetting to erroneously include minus signs; the data have been corrected in this version of the article.]
OI transfer and leadership behaviour 427
.10 .75
.51 .21
-.05 .14 **
.34 .07
**
20448325, 2012, 2, Downloaded from https://bpspsychub.onlinelibrary.wiley.com/doi/10.1111/j.2044-8325.2011.02044.x by University Of Texas-Arlington, Wiley Online Library on [14/09/2023].Across three studies, van Dick et al. (2007) found support for the proposed link between leader OI and employee OI. Several studies have replicated these findings in a wide range of organizational contexts including pharmaceutical companies, retail organizations, and schools (Lichtenstein, Netemeyer, & Maxham III, 2010; Schuh, Egold, & van Dick, 2011; Wieseke, Ahearne, Lam, & van Dick, 2009).See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

Transformational leadership
We used Bass and Avolio's (1995) Multifactor Leadership Questionnaire 5X short scale to assess employees' perceptions of TFL.DOI:10.1111/j.2044-8325.2011.02044.x

422 Sebastian C. Schuh et al.
o Furthermore, our findings suggest that this transfer of OI is not restricted to leader-follower links in western, individualistic societies but also applies to China, characterized by more collectivistic values.OI, organizational identification; TFL, transformational leadership; OCB, organizational citizenship behaviour; L_age, leader age; F_age, follower age; L_gender, leader gender; F_gender, follower gender; ~R2, pseudo-R2; CI, confidence interval.*
421
The
British Psychological Society
www.wileyonlinelibrary.com
Correspondence concerning this article should be addressed to Sebastian C. Schuh, Department of Psychology and Sports Sciences, Goethe University Frankfurt, Kettenhofweg 128, 60054 Frankfurt am Main, Germany (e-mail: [email protected]).The hypothesized two-factor measurement model yielded an acceptable level of fit ( 2 = 320.0, df = 76; comparative fit index (CFI) = .91; standardized root mean square residual (SRMR) = .07) and fitted the data significantly better than the one-factor model combining both variables(2 =6.1;df =1,p<.05).See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

428 Sebastian C. Schuh et al.
In Step 1 of the joint-significance test, leader OI was significantly related to TFL.[Corrections added on 14 December 2011 after initial online publication on 7 November 2011: some data errors have been corrected in this version of the article.]
.02 -.04 -.06 -.02 - .05
-.18+ - -.73* * * .77***
-.25* * -.07 -.29* * -.12 -.35* * * -.12 -.05 -.13
-.03 .12+ .15* -.08
-.20* -.27* * -.10
.50* * * .21* .20*
20448325, 2012, 2, Downloaded from https://bpspsychub.onlinelibrary.wiley.com/doi/10.1111/j.2044-8325.2011.02044.x by University Of Texas-Arlington, Wiley Online Library on [14/09/2023].See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

432 Sebastian C. Schuh et al.
MacKinnon, D. P., Lockwood, C. M., Hoffman, J. M., West, S. G., & Sheets, V. (2002).Hence, the results support our hypothesis that TFL mediates the link between leader OI and employee OI.
Supplemental analyses
To determine the robustness of our findings, we calculated a multivariate outlier analysis.However, in line with theory (e.g., upper echelon theory, Hambrick & Mason, 1984) and previous research (e.g., Schneider, Ehrhart, Mayer, Saltz, & Niles-Jolly, 2005), we believe that the general direction of influence in organizations follows the direction of formal power with leaders affecting followers.To ensure that both measures collected from followers (i.e., perceptions of TFL and follower OI) were distinct, we conducted confirmatory factor analyses.See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

430 Sebastian C. Schuh et al.
for this assumption.Since leaders are an important factor in shaping employees' attitudes and behaviours, van Dick et al. (2007) further posited that leaders' identification with the organization would influence their followers' OI. Higher levels of employee OI should, in turn, result in stronger efforts on behalf of the organization.This holds especially true given that the cross-cultural leadership research has yielded mixed results for the applicability of western leadership models to other cultural settings (Kirkman, Chen, Farh, Chen, & Lowe, 2009).The average intraclass correlation (ICC 1), which indicates the variability in participants' responses based on group membership, were .26 and .13 for TFL and employee OI, respectively (Bliese, 2000).See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

426 Sebastian C. Schuh et al.
Table 1.Again, we followed the translation procedures outlined by Brislin (1980) to
20448325, 2012, 2, Downloaded from https://bpspsychub.onlinelibrary.wiley.com/doi/10.1111/j.2044-8325.2011.02044.x by University Of Texas-Arlington, Wiley Online Library on [14/09/2023].The hypothesized three-factor measurement model yielded an acceptable fit (2 = 395.2, df = 206; CFI = .94; SRMR = .09) and fitted the data significantly better than all possible two-factor models (min.Across two studies, we found consistent evidence
OI transfer and leadership behaviour 429
20448325, 2012, 2, Downloaded from https://bpspsychub.onlinelibrary.wiley.com/doi/10.1111/j.2044-8325.2011.02044.x by University Of Texas-Arlington, Wiley Online Library on [14/09/2023].Practitioner Points
o Given that organizations are generally structured in a pyramid-shaped form, the present findings suggest that leader OI can be highly leveraged since it appears to cascade down the organizational hierarchy.Finally, research on affective commitment (which can be seen as a good proxy for OI; Riketta & van Dick, 2009) has shown that TFL is positively related to employees' commitment to the organization (Liao & Chuang, 2007) and to their supervisor (Felfe & Schyns, 2010).Controls
We controlled for similarity in age and gender between leaders and followers since similarity has a considerable impact on attraction and influence processes between human beings (Tsui & O'Reilly, 1989).To examine whether HLM was justified, we conducted a series of one-way ANOVAs
for all multi-level constructs using the corresponding unit variable (i.e., medical centre) as factor.In Step 2, when leader OI and TFL were simultaneously entered into the cross-level regression, TFL significantly predicted followers' level of OI. Finally, the confidence interval for the indirect effect (based on bootstrap analyses) did not include zero.See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

ensure equivalency of item meaning.Additionally, when leader OI and TFL were simultaneously entered into the cross-level regression, TFL significantly predicted followers' level of OI. Finally, the confidence interval of the bootstrap analyses did not include zero.The present studies seek to contribute to this literature in several ways: first, despite the cumulative evidence for the leader- follower OI transfer, none of the previous studies have examined the behavioural links that mediate this process.Second, given the growing power of collectivistic Asian countries such as China, we believe it is highly warranted to examine the generalizability of western concepts and models in these cultures.Additionally, these results contribute to a growing body of literature, which suggests the usefulness of the OI concept in collectivistic societies (e.g., Gautam, van Dick, Wagner, Upadhyay, & Davis, 2005).Potentially, the primary limitation concerns the cross-sectional design we had to rely on. Hence, it could be argued that the influence between leaders and followers is reciprocal instead of unidirectional.Given that globalization and increasing individual mobility appear to erode individual-organization ties, it seems essential that organizations actively counteract this development and strengthen the bond with their employees.Hence, measures targeted at increasing OI at the leadership level (e.g., refraining from short-term contracts and emphasizing the value of long-term work relationships; Johnson & Ashforth, 2008) can be an effective way to foster the bond between an organization and its employees.See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

References
Bass, B. M. (1985).Linking leader OI, leadership behaviour, and employee outcomes
OI has been defined as the 'perception of oneness with or belongingness to the organization' (Mael & Ashforth, 1992, p. 104).Drawing on the literature of transformational and charismatic leadership, van Dick et al. (2007) suggested that leaders high in OI are likely to engage in TFL behaviours that benefit the organization and its members.In another sample of bank employees
OI transfer and leadership behaviour 423
20448325, 2012, 2, Downloaded from https://bpspsychub.onlinelibrary.wiley.com/doi/10.1111/j.2044-8325.2011.02044.x by University Of Texas-Arlington, Wiley Online Library on [14/09/2023].We found significant between-group differences for all variables, which indicated that hierarchical modelling was warranted (Hofmann, Griffin, & Gavin, 2000).OI transfer and leadership behaviour 425
20448325, 2012, 2, Downloaded from https://bpspsychub.onlinelibrary.wiley.com/doi/10.1111/j.2044-8325.2011.02044.x by University Of Texas-Arlington, Wiley Online Library on [14/09/2023].OI transfer and leadership behaviour 431
20448325, 2012, 2, Downloaded from https://bpspsychub.onlinelibrary.wiley.com/doi/10.1111/j.2044-8325.2011.02044.x by University Of Texas-Arlington, Wiley Online Library on [14/09/2023].The present studies extend the OI transfer model by examining the mediating role of transformational leadership (TFL) behaviour in the leader OI-follower OI link.However, in light of the fact that western societies account for only one-fourth of the world's population and given the growing power of collectivistic countries (e.g., China) in politics and economics, it appears important to examine the generalizability of this model to non-western societies.See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

Figure 1.In their self-concept based theory of leadership, Shamir et al. (1993) argued that the influence of transformational leaders is based on their ability to connect followers' self-definition to the mission and goals of the organization.The scale consisted of eight items, two items representing each of the behavioural dimensions of TFL ( = .88; sample item: 'My supervisor helps me to develop my strengths').In line with previous research (e.g., Kark et al., 2003), we did not measure 'individualized influence attributed', which has been criticized for measuring leadership impact rather than leadership behaviour (Yukl, 2010).See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

Table 2.Again, a series of ANOVAs for all multi-level constructs indicated significant between-group differences, which justified hierarchical modelling.To calculate the confidence interval of the indirect effect, we additionally performed a product-of-coefficient test based on unbiased standard errors (Taylor et al., 2008).Even though explaining mediating processes (i.e., understanding the why behind phenomena) is a central objective of organizational research, testing mediating links which span more than one variable is surprisingly rare.Nevertheless, we would like to encourage future, longitudinal research to examine the relative strengths of potential reciprocal effects between leaders and followers.Acknowledgements
The present research was supported by a fellowship of the Alexander von Humboldt- Foundation awarded to Xin-an Zhang.Centering decisions in hierarchical linear models: Theo-
retical and methodological implications for organizational science.Received 26 March 2011; revised version received 29 June 2011
20448325, 2012, 2, Downloaded from https://bpspsychub.onlinelibrary.wiley.com/doi/10.1111/j.2044-8325.2011.02044.x by University Of Texas-Arlington, Wiley Online Library on [14/09/2023].See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons LicenseAccordingly, recent years have witnessed a growing interest in the factors that foster or hinder employee OI. One promising approach is the investigation of leadership influence.The self-concept based theory of leadership (Shamir, House, & Arthur, 1993) suggests that effective leaders are particularly adept at fostering employees' identification with the collective.Based on this notion, van Dick, Hirst, Grojean, and Wieseke (2007) proposed a transfer model of OI that links leader OI to follower OI and, in turn, to employee citizenship behaviour.The present research addresses both of these research gaps by examining a three-path mediational model that links leader OI and employee behaviour through the intervening role of TFL and employee OI (see Figure 1).20448325, 2012, 2, Downloaded from https://bpspsychub.onlinelibrary.wiley.com/doi/10.1111/j.2044-8325.2011.02044.x by University Of Texas-Arlington, Wiley Online Library on [14/09/2023].The authors identified four routes through which leader OI may influence leadership behaviour: first, leaders who identify with the organization are motivated to increase its status and, therefore, are likely to exert themselves for the organization.According to this view, TFL behaviours increase the salience of the organizational identity and, thus, activate followers' self-definition in terms of the organization.For instance, surveying 76 bank managers and 888 employees, Kark et al. (2003) found that TFL fostered followers' identification with the organization.STUDY 1
Procedure
We conducted Study 1 in cooperation with several specialist medical centres in the area of Frankfurt, Germany (e.g., centres for physiotherapy, speech therapy, and psychotherapy).20448325, 2012, 2, Downloaded from https://bpspsychub.onlinelibrary.wiley.com/doi/10.1111/j.2044-8325.2011.02044.x by University Of Texas-Arlington, Wiley Online Library on [14/09/2023].To test our mediational hypothesis, we applied the joint-significant test recommended by MacKinnon, Lockwood, Hoffman, West, and Sheets (2002).Since the data of this study was hierarchical in nature with followers being nested in teams, we applied hierarchical linear modelling (HLM).Additionally, TFL significantly predicted employee OI. These findings were corroborated by the results of the hierarchical model (see Table 2).OI, organizational identification; TFL, transformational leadership; OCB, organizational citizenship behaviour.As in Study 1, we conducted confirmatory factor analyses to ensure that all follower variables were distinct.Therefore, in addition to Step 1 and 2 of the joint-significance test, we regressed employee OCB on employee OI while controlling for all previous predictors of the mediation model (see Table 2).These findings were corroborated by the results of the HLM analyses (see Table 2): as in Study 1, leader OI was significantly related to employees' perceptions of TFL.GENERAL DISCUSSION
Past research has yielded consistent evidence for the leader-follower identity transfer model, which proposes a downstream flow of OI from leaders to followers.However, such analyses appear to be very beneficial to our understanding of organizational phenomena since many effects seem to be transmitted by more than one intervening variable (Taylor et al., 2008).20448325, 2012, 2, Downloaded from https://bpspsychub.onlinelibrary.wiley.com/doi/10.1111/j.2044-8325.2011.02044.x by University Of Texas-Arlington, Wiley Online Library on [14/09/2023].Does the transactional-transformational leadership paradigm transcend organi-
zational and national boundaries?In K. J. Klein & S. W. J. Kozlowski (Eds.), Mul- tilevel theory, research, and methods in organizations (pp.Individual power distance orientation and follower reactions to transformational leaders: A cross-level, cross-cultural examination.Organizational identification: A meta-analysis.2.3.4.5.6.7.8.9.