INTRODUCTION
Organizational socialization is the process by which an individual acquires the attitudes, behaviors and knowledge he or she needs to participate as an organization member (Van Maanen & Schein, Reference Van Maanen and Schein1979). This process is very important because of its enduring impact on newcomers’ attitudes and behaviors, and its function in maintaining the goals of the organization and its culture (Bauer, Morrison, & Callister, Reference Bauer, Morrison and Callister1998; Morrison, Reference Morrison2002a). Organizations thus try to use socialization tactics to help newcomers transfer from ‘outsiders’ to ‘insiders.’ In this process, the newcomers do not play a passive role. Newcomers’ proactive behavior can help them reduce uncertainty and adapt themselves well to the new environment (Ashford & Black, Reference Ashford and Black1996). When they enter a new organization, newcomers are trying to acquire feedback about their work, information about their new roles as well as to understand the organization’s social network in order to minimize uncertainty and help themselves to blend into the new environment. The interactionist perspective would seek to integrate these two sides by examining how a newcomer’s attempts at self-socialization work in tandem with the organization’s attempts at socialization to influence the adjustment outcomes (Griffin, Colella, & Goparaju, Reference Griffin, Colella and Goparaju2000; Gruman, Saks, & Zweig, Reference Gruman, Saks and Zweig2006; Li, Harris, Boswell, & Xie, Reference Li, Harris, Boswell and Xie2011; Reichers, Reference Reichers1987). In this connection, it is the proactive socialization practices of a supervisor that are often the most available and helpful for a newcomer (Anderson & Cooper-Thomas, Reference Anderson and Cooper-Thomas1996; Cooper-Thomas, Anderson, & Cash, Reference Cooper-Thomas, Anderson and Cash2012). As the newcomer tries proactively to adapt to the new organization, daily interactions with the supervisor can help him or her acquire information and ‘learn the ropes’ (Morrison, Reference Morrison2002a).
One major mission in the existing study is the use of interactionist perspective as an overarching theory to develop a cross-level framework that integrates the transformational leadership and uncertainty reduction literature to examine the impact of transformational leadership on the newcomer’s information seeking and adjustment in a work group context. The interactionist perspective as employed here emphasizes the idea that the newcomers’ acculturation stems from interaction with the supervisor, who plays the role of a socialization agent. Kozlowski and Bell (Reference Kozlowski and Bell2003) proposed that the impact of a supervisor on the newcomer adjustment and their mutual proaction have not been sufficiently explored in published academic work. Although ‘leader influence’ has been treated in socialization research and the supervisor’s critical effect on newcomer socialization has been verified (Cooper-Thomas & Anderson, Reference Cooper-Thomas and Anderson2006; Kammeyer-Mueller & Wanberg, Reference Kammeyer-Mueller and Wanberg2003), it is not clear to what extent the leadership style of the socialization agent influences his or her actions. To fill the void, this study focused on the transformational leadership style because it has been extensively shown that it has a positive effect on employees’ attitudes and behaviors (Judge & Piccolo, Reference Judge and Piccolo2004). This is an attempt to learn whether transformational leadership, as one of the characteristics of the socialization agent, could show the proaction to the newcomers and influence their proactive behaviors. Based on the research of Bauer’s group (Bauer, Bodner, Erdogan,Truxillo, & Tucker, Reference Bauer, Bodner, Erdogan, Truxillo and Tucker2007), we took the information seeking of newcomers as a typical proactive behavior to observe their proactivity during the process of socialization.
Further, information seeking has been shown to have a positive effect on newcomers’ adjustment outcomes in research spanning almost 2 decades (Cooper-Thomas, Anderson, & Cash, Reference Cooper-Thomas, Anderson and Cash2012; Morrison, Reference Morrison1993b). However, scholars have noticed that the antecedents of information seeking are not completely predictable and there is a need to identify these antecedents (Cooper-Thomas, Anderson, & Cash, Reference Cooper-Thomas, Anderson and Cash2012; Wanberg & Kammeyer-Mueller, Reference Wanberg and Kammeyer-Mueller2000). To address this gap, this study was designed to examine how transformational leadership influences information seeking, which in turn explains how and why a transformational leader as the socialization agent affects newcomers’ adjustment outcomes.
Accordingly, this study integrated two types of mechanisms and examined the sequential relationship between multiple mediators to explore the knowledge of the work group socialization. More specifically, it tested the proposition that transformational leadership motivates a newcomer to become more proactive and thus to demonstrate more information seeking behavior. That proactivity (information seeking) was shown to be positively related to proximal adjustment outcomes (role clarity and social integration) and also to distal adjustment outcome (affective organizational commitment), which are the most studied adjustment outcomes in the socialization literature (Bauer et al., Reference Bauer, Bodner, Erdogan, Truxillo and Tucker2007; Kammeyer-Mueller & Wanberg, Reference Kammeyer-Mueller and Wanberg2003).
Finally, the supervisor and the work group structure have previously been identified as the contextual varibles of work group that can influence the newcomer socialization (Ashforth, Sluss, & Harrison, Reference Ashforth, Sluss and Harrison2007; Moreland, Levine, & McMinn, Reference Moreland, Levine and McMinn2001), and characteristics of the structure influence the frequency of interaction between a supervisor and a newcomer (Reichers, Reference Reichers1987). However, prior studies have rarely examined how the multiple contextual variables which can influence newcomer socialization to jointly influence the newcomer socialization process (exceptions include the studies by Ashforth, Sluss, & Saks, Reference Ashforth, Sluss and Saks2007; Bauer, Morrison, & Callister, Reference Bauer, Morrison and Callister1998; Cooper-Thomas & Anderson, Reference Cooper-Thomas and Anderson2006; Kammeyer-Mueller & Wanberg, Reference Kammeyer-Mueller and Wanberg2003; Li et al., Reference Li, Harris, Boswell and Xie2011). Extending that prior research, this study examined how work group structure moderates the relationship between transformational leadership and information seeking.
In sum, the goal of the current study was to further shed additional light on the socialization process by defining how the contextual variables of a work group influence newcomer adjustment. This involved by linking transformational leadership to information seeking that simultaneously relates to role clarity and social integration, which, in turn, both relate to affective organizational commitment. Meanwhile, the moderating effect of work group structure on the relationship between transformational leadership and information seeking was examined in detail. Figure 1 visually presents the overall hypothesized framework.
Hypothesized relationships
THEORETICAL BACKGROUND AND HYPOTHESES
As a socialization agent, a leader can offer his or her followers basic values, beliefs and attitudes which will lead them to align themselves with the organization’s collective interests (Podsakoff, MacKenzie, Moorman, & Fetter, Reference Podsakoff, MacKenzie, Moorman and Fetter1990). Previous research results have supported arguments that transformational leadership has a positive effect on employee’s behavior and attitudes, including those of both newcomers and employees who have already worked in an organization for years (e.g., Judge & Piccolo, Reference Judge and Piccolo2004; Kark, Shamir, & Chen, Reference Kark, Shamir and Chen2003; Song, Tsui, & Law, Reference Song, Tsui and Law2009). The impact of transformational leadership on newcomers is likely, however, to be more significant. The main features of transformational leadership have demonstrated that the supervisor can act as a socialization agent by: articulating a vision of the organization’s future; acting as a role model consistent with that vision, expressing clear performance expectations, convincing a newcomer of the importance of the group’s goals, and generally providing individual support and intellectual stimulation (Chen & Farh, Reference Chen and Farh1999; Wang, Law, Hackett, Wang, & Chen, Reference Wang, Law, Hackett, Wang and Chen2005).
This study examining transformational leadership at group level because the behaviors of transformational leadership are directed at the entire work unit and help the group share similar perceptions (Kirkman, Chen, Farh, Chen, & Lowe, Reference Kirkman, Chen, Farh, Chen and Lowe2009). Although the dyadic processes of transformational leadership will create different perceptions of their leader’s transformational leadership behaviors among the group members (Wang & Howell, Reference Wang and Howell2010), other transformational leadership behavior influences the members as a group (e.g., acting as a role model). Many previous empirical studies have treated it as a group-level phenomenon (e.g., Bono & Judge, Reference Bono and Judge2003; Kirkman et al., Reference Kirkman, Chen, Farh, Chen and Lowe2009). Newcomers’ role clarity, social integration and affective organizational commitment of course differ from individual to individual and evolve over time. Information seeking varies similarly (Morrison, Reference Morrison2002b). Transformational leadership is thus best treated at the group level, but information seeking and adjustment outcomes on the individual level.
Transformational leadership and newcomer adjustment outcomes
Role clarity, social integration and affective organizational commitment are important adjustment outcomes. Role clarity refers to the newcomer’s level of understanding of his or her job and the firm’s expectations. Social integration refers to the newcomer’s integration into his or her new work group. Affective organizational commitment refers to the newcomer’s emotional attachment to, identification with and involvement in the organization (Allen & Meyer, Reference Allen and Meyer1990; Ashford & Black, Reference Ashford and Black1996).
When a newcomer joins an organization it may be quite difficult for them to understand clearly their new working role. It may also be challenging for them to accept the organization’s culture completely and make a commitment to the new organization within a short span of time. At that moment transformational leader can serve as a counselor, providing individual support to the newcomer and offering valuable advice about the jobs, helping to solve problems they encounter. The transformational leader provides strong support, builds confidence and encourages the newcomer to fit into the organization’s culture and adopt the group’s goals. The transformational leader’s extensive working experience, expertise and skills quite possibly enable him or her to become a role model the newcomer and for the group as a whole. As the newcomer achieves role clarity and social integration, he or she transforms from an outsider to an insider guided by the transformational leader. Transformational leadership is also expected to result in affective organizational commitment, and previous studies provide strong evidence to this proposition (e.g., Bono & Judge, Reference Bono and Judge2003; Podsakoff, MacKenzie, & Bommer, Reference Podsakoff, MacKenzie and Bommer1996). Therefore, it is reasonable to predict that transformational leadership has a positive impact on newcomer adjustment outcomes.
Hypothesis 1: Transformational leadership will positively influence the newcomer’s role clarity, social integration and affective organizational commitment.
The mediating role of information seeking
Information seeking is newcomer’s proactive behavior which can not only help them reduce their uncertainty and adapt to a new environment, but also facilitate their coming up to speed (Cooper-Thomas, Anderson, & Cash, Reference Cooper-Thomas, Anderson and Cash2012; Morrison, Reference Morrison2002b). In the socialization process, a transformational leader can directly stimulate proactive behavior of the newcomer (Den Hartog & Belschak, Reference Den Hartog and Belschak2012). The charisma of a transformational leader can offer a role model, leading the newcomer to trust the leader’s ability and willingly seek information from him or her. Individualized consideration from transformational leader to newcomers will give them personal attention, as the newcomers will be treated individually. These actions enable the formation of a close relationship between the leader and the employee, which builds an effective channel for information transmission. Research has revealed that information seeking may incur a social cost, which may decrease the proactivity of a newcomer (Morrison, Reference Morrison1993b). However, high expectations and valuable advice from the transformational leader can tend to drive newcomers to transcend their self-interest for the collective welfare. This could further encourage newcomers to seek information. Meanwhile, when the transformational leader realizes a newcomer is willing to follow and to seek information for collective not individual needs, they will tend to give the newcomer more information (Allen & Rush, Reference Allen and Rush1998). Madzar (Reference Madzar2001) has shown that a transformational leader can help employees reduce the degree of subjective inference when faced with an uncertain situation, the feeling of ‘losing face’ and the cost of obtaining information. A transformational leader can also influence the newcomer’s mode of thinking by repositioning some previous problems in a new environment and with new perspective, thus inspiring a more creative solution. Scholars also argue that a transformational leader will build a safe climate through the providing feedback and advice. This can reduce the cost of seeking information (Morrison, Reference Morrison2002b; Morrison & Vancouver, Reference Morrison and Vancouver2000).
Information seeking as one of the antecedents of newcomer adjustment gives a newcomer information about the new environment, reducing the uncertainty, making sense of the new job and helping the newcomer to adapt to the job more quickly (Bauer et al., Reference Bauer, Bodner, Erdogan, Truxillo and Tucker2007; Morrison, Reference Morrison2002a). Obtaining information about their working role and the organization is all part of socialization, because the information tells them what they should be doing in their new job and work environment, as well as how they can be accepted and meet the others’ role expectations through appropriate behaviors (Ashford & Black, Reference Ashford and Black1996; Morrison, Reference Morrison1993a). The information also helps newcomers understand the organization’s culture and how their social behavior is being evaluated. It clarifies what their coworkers consider to be appropriate and inappropriate attitudes and behaviors, enabling them to adjust their behaviors according to the work group’s norms (Morrison, Reference Morrison1993a). Therefore, seeking information initially can help them understand what is needed to function on the job (role clarity), and also what attitudes and behaviors will be accepted by the others (social integration) (Bauer et al., Reference Bauer, Bodner, Erdogan, Truxillo and Tucker2007). As one of the antecedents of newcomer adjustment, information seeking is effective in helping newcomers adapt to new organizations, which improves adjustment outcomes (Ashforth, Sluss, & Saks, Reference Ashforth, Sluss and Saks2007; Bauer et al., Reference Bauer, Bodner, Erdogan, Truxillo and Tucker2007; Gruman, Saks, & Zweig, Reference Gruman, Saks and Zweig2006; Morrison, Reference Morrison1993a; Saks, Gruman, & Cooper-Thomas, Reference Saks, Gruman and Cooper-Thomas2011). Briefly, we propose the links of transformational leadership with information seeking and proximal adjustment outcomes.
Hypothesis 2: Information seeking mediates the positive relationships of transformational leadership on the newcomer’s role clarity and social integration.
The mediating role of role clarity and social integration
Role clarity gives a newcomer a sense of direction and purpose on the job and also promotes task participation. This enables a newcomer to show his or her commitment to the organization. Conversely, role ambiguity leads to stress and potentially to burnout. Newcomers who are not completely clear about their role may thus develop less commitment to the organization (Saks, Uggerslev, & Fassina, Reference Saks, Uggerslev and Fassina2007). Social integration enables the newcomer to establish a situational identity and facilitates social interaction with other group members. It could help the newcomer obtain social support from other members in the group, which could enhance affective organizational commitment. Information seeking is also expected to result in high affective organizational commitment because the information will guide newcomers’ behavior and reduce entry uncertainty, and also lead to the formation of relationships and social networks (Gruman, Saks, & Zweig, Reference Gruman, Saks and Zweig2006). In addition, numerous studies have confirmed that information seeking has a positive effect on a newcomer’s organizational commitment (e.g., Bauer & Green, Reference Bauer and Green1998; Bauer et al., Reference Bauer, Bodner, Erdogan, Truxillo and Tucker2007). Hence, we hypothesize:
Hypothesis 3: Role clarity and social integration both mediate the relationship between information seeking and affective organizational commitment.
Hypothesis 2 predicts that if Hypothesis 1 is true, then information seeking will mediate the observed relationship between transformational leadership and the proximal adjustment outcomes. Similarly, Hypothesis 3 predicts that the relationship between information seeking and affective organizational commitment will be mediated by role clarity and social integration simultaneously. By influencing information seeking, transformational leadership presumably becomes an indirect antecedent to the proximal adjustment outcomes, which then translate into the distal outcome. Based on Hypotheses 2 and 3, we predict:
Hypothesis 4: Transformational leadership is indirectly related to the newcomer’s affective organizational commitment through the mediating role of information seeking and, in turn, through role clarity and social integration simultaneously.
The moderating role of work group structure
Work group structure is one important contextual factor that would be expected to either facilitate or constrain a leader’s ability to demonstrate transformational leadership. The work group’s structure also affects its members’ demands and choices, as well as their reactions to the transformational leader (Green, Anderson, & Shivers, Reference Green, Anderson and Shivers1996; Walter & Bruch, Reference Walter and Bruch2010). Other scholars have made similar proposals. For example, Antonakis and Atwater (Reference Antonakis and Atwater2002) have argued that the characteristics of the structure can explain how the followers perceive their leader. Therefore, the joint effects of work group structure and the leader typically determine how the work group influences information exchange between a transformational leader and a newcomer.
Work group structure is often mandated by organizational rules, but it influences patterns of decision making, communication among work group members and power relationships (Ambrose & Schminke, Reference Ambrose and Schminke2003; Donaldson, Reference Donaldson1996). Scholars have defined organic and mechanistic structures as typical of many work groups (Ambrose & Schminke, Reference Ambrose and Schminke2003; Slevin & Covin, Reference Slevin and Covin1997; Sun & Pan, Reference Sun and Pan2011). Organic structures are more flexible, loose and decentralized. Their communication channels are open and more flexible, power is decentralized and formal lines of authority are less clear (Ambrose & Schminke, Reference Ambrose and Schminke2003). In contrast, a mechanistic structure shows the characteristics of a rigid, tight, traditional bureaucracy. Power is centralized, communication follows rigid hierarchical channels, managerial styles and job descriptions are uniform.
In an organic structure, people will have more face-to-face communication and interpersonal interactions (Ambrose & Schminke, Reference Ambrose and Schminke2003). This may include interactions between a transformational leader and newcomers. In addition, flexible and open communication channels may facilitate a transformational leader’s offering information to newcomers. In a more mechanistic structure, communication channels are limited, information is transmitted through hierarchical channels and the rules of the organization may define the extent of information exchange. Hence, though the transformational leader will offer some information in integrating the newcomer and the newcomer may seek information initially, the amount, content and significance of the information may decrease over time. This may constrain information seeking and decrease the frequency of interaction between transformational leader and the newcomers. This is how structural form may strengthen or hinder the influence of transformational leadership on a newcomer’s proactivity.
Hypothesis 5: Work group structure moderates the strength of the positive relationship between transformational leadership and information seeking, such that the relationship will be stronger in an organic structure than in a mechanistic structure.
METHOD
Sample and procedure
The dyadic data used to test our hypothesis have been collected from seven five-star hotels in south China. New full-time employees evaluate the transformational leadership of their supervisor. They were also asked about the characteristics of their work group’s structure, and about their own role clarity, social integration and affective organizational commitment. The data were collected in four waves as the respondents completed their first 3 to 5 months of full-time work in the organization, with each wave designed to correspond with the presumed causal sequences proposed in Figure 1. At Time 1 the new employees reported their personal information (e.g., organizational tenure) and evaluated their immediate supervisor’s leadership and the characteristics of their work group. We asked participants to provide their employee identification so that we could match data at different time periods and later request their supervisor’s evaluation. We also provided assurances of strict confidentiality. A total of 500 surveys were distributed and 378 completed surveys were returned, a response rate of 75%. At Time 2 (about 2 weeks later), the new employees who completed the Time 1 surveys completed a second survey designed to assess their degree of understanding of their role and their degree of feeling accepted by the others in their work group. A total of 298 complete surveys were returned, a response rate of 79%. At Time 3 (about 2 weeks after Time 2) the new employees who completed the Time 2 surveys were given a final survey that assessed their commitment to their organization. A total of 226 complete surveys were returned, a response rate of 76%. At Time 4 (about 1 week later than Time 3), each newcomer’s supervisor was asked to evaluate how frequently the newcomer was seeking information. Each supervisor was requested to rate at least three new employees (the range was from three to five). This data collecting process helped to decrease common method variance and explore any causality among the variables (Podsakoff, MacKenzie, Lee, & Podsakoff, Reference Podsakoff, MacKenzie, Lee and Podsakoff2003). In order to motivate participation, a ¥60 gift voucher was offered to those who finished the whole sequence of surveys.
The final usable sample comprised 53 supervisors and 212 new employees (the average group size was 4). Among the 53 supervisors, 76% were male, the mean age was 32.6 years, and the mean organizational tenure was 8 years. Among the new employees, 52% were male and the mean age was 22.8 years, and the mean organizational tenure was 3 months. Over 84% held at least a college diploma.
Measures
Likert-type scales were used to make the assessments. Except for the transformational leadership scale, the original scales are all originally prepared in English, and then translated into Chinese following the standard translation and back-translation procedures (Brislin, Reference Brislin1986).
Transformational leadership style was assessed by 23 items adopted from the study by Chen and Farh (Reference Chen and Farh1999). This scale has been validated in previous studies in a Chinese context (e.g., Wang et al., Reference Wang, Law, Hackett, Wang and Chen2005). A sample item is ‘Articulates a compelling vision of the future for our group.’ The 5-point response scale ranged from 1=‘strongly disagree’ 5=‘strongly agree.’ Fit indexes fell within an acceptable range (χ2=397.28, p<.01, df=224; RMSEA=0.08; CFI=0.92; TLI=0.90).
Information seeking was rated by the newcomers’ supervisors using Chan and Schmitt’s (Reference Chan and Schmitt2000) 8-item scale. Some of the items were reworded slightly (e.g., changing the referent from ‘you’ to ‘his or her’) because information seeking was assessed by the supervisor rather than using the newcomers’ self-reports. A sample item is, ‘How frequently does he or she ask you for information about how his or her performance will be evaluated.’ The responses were on a 7-point scale covering 1=‘never’ 2=‘once a month’ 3=‘a few times a month’ 4=‘once a week’ 5=‘a few times a week’ 6=‘once a day’ 7=‘a few times a day.’ The fit indexes again fell within an acceptable range (χ2=46.28, p<.01, df=20; RMSEA=0.04; CFI=0.95; TLI=0.94).
Work group structure was measured by 7 items. These were developed by Slevin and Covin (Reference Slevin and Covin1997), and later validated in Ambrose and Schminke’s research (Ambrose & Schminke, Reference Ambrose and Schminke2003). Sample items are, ‘A strong insistence on a uniform managerial style throughout the business unit’ versus ‘Managers’ operating styles allowed to range freely from the very formal to the very informal’ and ‘Tight formal control of most operations by means of sophisticated control and information systems’ versus ‘Loose, informal control; heavy dependence on informal relationships and the norm of cooperation for getting things done.’ Newcomers indicated along a 5-point scale the degree to which paired statements described the structure of their work group. Items were scored such that higher values represented a more organic structure. Here too the fit indexes fell within an acceptable range (χ2=35.93, p<.05, df=14; RMSEA=0.06; CFI=0.94; TLI=0.92).
Role clarity was measured with 10 items from a study by Morrison (Reference Morrison2002a). A sample item is, ‘I feel certain how much authority I have in my job.’ Responses were on a 5-point scale ranging from 1=‘strongly disagree’ 5=‘strongly agree.’ Fit indexes fell within an acceptable range (χ2=92.37, p<.01, df=35; RMSEA=0.05; CFI=0.93; TLI=0.92).
Social integration was measured with 7 items, also from the study by Morrison (Reference Morrison2002a). A sample item is, ‘My coworkers seem to accept me as one of them.’ Responses were on a 5-point scale ranging from 1=‘strongly disagree’ 5=‘strongly agree.’ Fit indexes here too fell within an acceptable range (χ2=29.58, p<.05, df=14; RMSEA=0.09; CFI=0.91; TLI=0.90).
Affective organizational commitment was measured with 6 items from the study by Francesco and Chen (Reference Francesco and Chen2004). A sample item is, ‘I would be very happy to spend the rest of my career with this organization.’ Responses were made on a 5-point scale ranging from 1=‘strongly disagree’ 5=‘strongly agree.’ These fit indexes too fell within an acceptable range (χ2=21.28, p<.01, df=9; RMSEA=0.06; CFI=0.94; TLI=0.93).
Control variables
Newcomers’ demographic variables such as the organizational tenure, age, gender and education were controlled in this research because these demographic variables have potential effects on newcomers’ reactions to the socialization process (Kammeyer-Mueller, Livingston, & Liao, Reference Kammeyer-Mueller, Livingston and Liao2011; Kim, Cable, & Kim, Reference Kim, Cable and Kim2005; Li et al., Reference Li, Harris, Boswell and Xie2011). Newcomers with an organization a relatively short time may differ from more senior employees. Tenure is often used as a control variable in socialization research (Allen, Reference Allen2006). The other demographic variables may also represent life experiences relevant to the socialization process (Bauer, Morrison, & Callister, Reference Bauer, Morrison and Callister1998).
Level of analysis
Kozlowski and Klein (Reference Kozlowski and Klein2000) argued that the key point of specifying the level of analysis is that at which researchers are assuming variables will operate. As discussed above, transformational leadership could impact all group members who may have a similar perception about the leadership style within one group (Kirkman et al., Reference Kirkman, Chen, Farh, Chen and Lowe2009). Transformational leadership was therefore examined at the group level in this research. Similarly, work group structure was treated at the group level because the group’s structure is a shared phenomenon for the group’s members (Ambrose & Schminke, Reference Ambrose and Schminke2003). Following the method of LeBreton and Senter (Reference LeBreton and Senter2008), the values of r wg(j), ICC(1) and ICC(2) were calculated. For transformational leadership, the average of the within-group agreement measure with respect to the median r wg(j) value was 0.87, and there was significant between-group variance: F (52, 159)=1.43, p<.05. The ICC(1) value was 0.45 and ICC(2) value was 0.76, both within the acceptable range of values (Bliese, Reference Bliese2000). For work group structure, the median r wg(j) value was 0.89, ICC(1) value was 0.42 and ICC(2) value was 0.74, F (52, 159)=2.75, p<.01. Thus, these results provide sufficient evidence to aggregate transformational leadership and work group structure to the group level (Kozlowski & Hattrup, Reference Kozlowski and Hattrup1992; Chen & Bliese, Reference Chen and Bliese2002).
Analysis strategy
Hierarchical linear modeling was used to test the hypothesis. Raw score variables were used to conduct the mediation test, without centering (Bliese, Reference Bliese2000). Although transformational leadership and work group structure were evaluated at the group level, information seeking, role clarity, social integration and affective organizational commitment were all assessed at the individual level. The hypotheses were tested using the mediation procedures in line with the four-step process of Kenny, Kashy, and Bolger (Reference Kenny, Kashy and Bolger1998).
RESULTS
Table 1 presents the means, standard deviations, correlations and reliabilities for all of the variables. All of the coefficient α’s ranged from 0.77 to 0.94 and showed high internal reliability. In line with the hypotheses, transformational leadership showed a statistically significant positive relationship with information seeking, role clarity, social integration and affective organizational commitment. Information seeking also showed a statistically significant positive relationship with role clarity, social integration and affective organizational commitment. Finally, role clarity and social integration were significantly and positively related to affective organizational commitment.
Means, standard deviations and intercorrelations among the variables
Notes. n=212; reliability coefficients for the scales are in parentheses along the diagonal. Transformational leadership and work group structure were calculated as group-level means and assigned back to members of the same group.
*p<.05; **p<.01.
Confirmatory factor analysis
Confirmatory factor analyses were performed with the aid of version 8.7 of the LISREL software package. To form the measurement models, three indicators were randomly created for each of work group structure, information seeking, role clarity, social integration and affective organizational commitment. Parceled indicators of transformational leadership were defined using its six dimensions as its indicators (Little, Cunningham, Shahar, & Widaman, Reference Little, Cunningham, Shahar and Widaman2002).The hypothesized six-factor model relating transformational leadership, work group structure, information seeking, role clarity, social integration and affective organizational commitment fitted the data well. The hypothesized six-factor model was compared with other alternative measurement models, and Table 2 shows the CFI results. They suggest that the hypothesized measurement model displays a better fit than each of the alternative models, thus providing support for the distinctiveness of the constructs in this study.
Comparison of measurement models
Notes. Model 1: transformational leadership and work group structure were combined into one factor; Model 2: role clarity and social integration were combined into one factor; affective organizational commitment and work group structure were combined into one factor; Model 3: role clarity, social integration and affective organizational commitment were combined into one factor; transformational leadership and work group structure were combined into one factor; Model 4: transformational leadership, information seeking and work group structure were combined into one factor; role clarity, social integration and affective organizational commitment were combined into one factor.
**p<.01.
Hypothesis tests
Hypothesis 1 proposes a direct relationship between transformational leadership and adjustment outcomes. Hypothesis 2 proposed that information seeking mediates the positive relationship of transformational leadership on role clarity and social integration. Tables 3 and 4 show the results of hypothesis testing, following the procedure described by Kenny, Kashy, and Bolger (Reference Kenny, Kashy and Bolger1998). Models 1, 4, 7 and 9 (see Tables 3 and 4) shown that transformational leadership was positively related to role clarity, social integration, information seeking and affective organizational commitment. Information seeking was positively related to role clarity and to social integration, as is shown in Models 2 and 5 (see Table 3). Role clarity and social integration were regressed on both transformational leadership and information seeking. As shown in Models 3 and 6 (Table 3), after controlling for information seeking, transformational leadership was not significantly related with role clarity or with social integration. So information seeking fully mediates the relationship between transformational leadership and role clarity, and between transformational leadership and social integration.
Hierarchical linear modeling results for hypotheses
Notes. Newcomers n=212, groups n=53. Entries are estimations of the fixed effects (γs) with robust standard errors.
a R 2 is based on the proportional reduction of levels 1 and 2 error variance resulting from predictors (Raudenbush & Bryk, Reference Raudenbush and Bryk2002).
M=model.
†p<.1; *p<.05; **p<.01.
Hierarchical linear modeling results for hypotheses
Notes. Newcomers n=212, groups n=53. Entries are estimations of the fixed effects (γs) with robust standard errors.
a R 2 is based on the proportional reduction of levels 1 and 2 error variance resulting from the predictors (Raudenbush & Bryk, Reference Raudenbush and Bryk2002).
M=model.
*p<.05; **p<.01.
Hypothesis 3 suggests that role clarity and social integration both mediate the relationship between information seeking and affective organizational commitment. Models 2 and 5 (see Table 3) show that information seeking was significantly related to both role clarity and social integration. Similarly, Model 10 (see Table 4) shows information seeking was significantly related to affective organizational commitment. Furthermore, Models 11 and 12 show that role clarity and social integration were significantly related to affective organizational commitment. Finally, Models 13 and 14 (see Table 4) reveal that role clarity and social integration both mediate the relationship between information seeking and affective organizational commitment.
Hypothesis 4 predicts that transformational leadership is indirectly related to affective organizational commitment through the mediating role of information seeking and, in turn, through role clarity and social integration simultaneously, as shown in Figure 1. Model 15 (see Table 4) shows that information seeking and role clarity were significantly related to affective organizational commitment; transformational leadership, however, was no longer significantly related to affective organizational commitment. Model 16 (Table 4) shows that information seeking and social integration were still significantly related to affective organizational commitment, but transformational leadership was not. These results indicate that Hypothesis 4 has been supported.
To test the significance of the indirect effect of transformational leadership on affective organizational commitment via information seeking and the proximal adjustment outcomes in series, a 95% bias-corrected confidence interval was constructed based on 10,000 bootstrap samples (Hayes, Preacher, & Myers, Reference Hayes, Preacher and Myers2011). The 95% confidence interval [0.15, 0.40] via social integration (transformational leadership→information seeking→social integration→ affective organizational commitment) did not include zero, and the 95% confidence interval [0.11, 0.37] via role clarity (transformational leadership→information seeking→role clarity→affective organizational commitment) also did not include zero. This is further evidence for a significant indirect effect passing through the two mediators, and it supports Hypothesis 4.
Hypothesis 5 propos a cross-level interaction effect between transformational leadership and work group structure on information seeking. To examine the cross-level interaction, a ‘mean-as-outcomes’ model was evaluated (Model 8). The interaction effect was significant (γ03=0.71, p<.05), and the interaction between transformational leadership and work group structure explained 3% (∆R 2=0.03) of the variance in information seeking. To further understand interaction results, the interaction effect was plotted based on the procedure given by Aiken and West (Reference Aiken and West1991). Figure 2 presents the results. Hypothesis 5 was supported.
Moderating effect of work group structure on the relationship between transformational leadership and information seeking
DISCUSSION
Based on the interactionist perspective, a relationship between transformational leadership and affective organizational commitment was predicted sequentially mediated by information seeking and the proximal adjustment outcomes. The results show that transformational leadership is indeed positively associated with information seeking, and with the proximal and distal outcomes. Furthermore, information seeking provides an explanation for the relationship between transformational leadership and adjustment outcomes. In addition, the results show that the joint effects of the two contextual factors can influence newcomers’ proactive behavior.
Theoretical implications
Taken together, these results extended the research of Reichers (1987). She argued that the characteristics of the socialization agent (e.g., a supervisor or coworker) and of the newcomer both could influence their interaction frequency. The results here show that the transformational leadership style, as a characteristic of the supervisor, can influence the frequency of interaction between the supervisor and a newcomer. These results extend the literature on socialization by revealing that the reason why transformational leaders have a positive effect on newcomer adjustment outcomes is that such leaders motivate the willful seeking of information by newcomers.
Most previous studies only explained how a newcomer’s proactivity influences their socialization. Few have sought to explain the antecedents of newcomers’ proactive socialization tactics. The results of this study offer some empirical support for an explanation of how and why newcomers demonstrate information seeking behavior in their socialization, answering a call of Cooper-Thomas, Anderson, and Cash (Reference Cooper-Thomas, Anderson and Cash2012).
This research also sheds some light on how the work group context influences newcomer adaption. A work group’s structure apparently moderates the relationship between transformational leadership and information seeking. An organic structure strengthens the positive effect of transformational leadership on information seeking, but a mechanistic structure weakens it. This result will question the effectiveness of transformational leader on information seeking in mechanistic structure. Transformational leaders may still, though, have significant influence on information seeking in a mechanistic structure because it comes from supervision structure that represents the frequency of leader–follower interaction (Antonakis & Atwater, Reference Antonakis and Atwater2002). For example, consideration as one of the typical behaviors of a transformational leader, which belongs to the supervision structure, makes dyad interacts not only on formal work but also on affective and non-task related basis. The proactive behaviors of a transformational leader will make him or her become more propinquity and create more social and task contact in the dyadic relationship (Napier & Ferris, Reference Napier and Ferris1993). Hence, newcomers are less likely to be isolated from the other group members by a mechanistic structure and a transformational leader will still be able to interact with newcomers. These results highlight the importance of the roles of the leader and the group in socialization, answering the call of several scholars (e.g., Ashforth, Saks, & Lee, Reference Ashforth, Saks and Lee1998; Kozlowski & Bell, Reference Kozlowski and Bell2003; Morrison, Reference Morrison2002b).
Finally, the findings verify the generalizability of socialization research in a Chinese cultural context. Globalization has made China an important participant in the world economy, which has led to the popularization of Western-style management in China.
Managerial implications
Effective socialization, by definition, helps new employees understand the significance of working well, helps them identify with the new organization and helps them more quickly accept the organization’s values. The findings in this study suggest that transformational leadership is effective in acculturating a new employee. Supervisors with a transformational leadership style are the best candidates for undertaking the role of socialization agent in a work group context. Therefore, organizations should make a transformational leader the socialization agent, even if this means more supervisors must be trained to develop that leadership style. Supervisors should be prepared to employ different tactics in managing new employees compared with more senior ones, because the new employees’ readiness for work, competence and psychological maturity are likely to be different.
In addition, these results verify that information is one critical factor that facilitates the adjustment of new employees. Acquiring useful information is important to reduce uncertainty and make sense of the new environment. Therefore, it is desirable for managers and the socialization agent to open various channels to provide sufficient information to new employees in a timely manner. That will help them adapt to the working environment more quickly, efficiently and effectively.
Work group structure should also be taken into consideration. An organic structure such as the autonomous team should help managers transmit information. New employees should be encouraged to seek information from their supervisors. Compliments and rewards are desirable for promoting proactivity.
Limitations and future research
In view of the dearth of socialization literature from an interactionist perspective, the current results must be viewed as preliminary and exploratory and their limitations must be borne in mind. First, this study was conducted within a single industry: Chinese hotels. Further research should extend these findings to other industries and be carried out in a Western culture in order to define the generalizability of the results.
Beyond that, this study did not consider the influence of individual characteristics such as a proactive personality or working experience. Therefore, a useful extension would be to include the newcomer’s personality and examine whether personality influences proactivity and the response to a transformational leader. Then, coworkers are another socialization agent, and also an obvious source of information (Kammeyer-Mueller & Wanberg, Reference Kammeyer-Mueller and Wanberg2003; Morison, Reference Morrison1993b). Further studies might fruitfully analyze to what extent the characteristics of coworkers influence a newcomer’s proactive information seeking.
The possibility of common method variance must also be acknowledged, since the new employees self-evaluated their commitment, the proximal adjustment outcomes, transformational leadership and the work group’s structure. The data were, however, collected at different points in time from new employees and their supervisors to reduce the effect of common method variance. Moreover, transformational leadership and work group structure were treated as group-level variables, which helped reduce possible common method variance by averaging responses within each group in the scoring (Kirkman et al., Reference Kirkman, Chen, Farh, Chen and Lowe2009). Taken together, these design features add some degree of confidence to our conclusions.
However, these results provide only a first step in understanding work group socialization, because only the influence of a transformational leader and work group structure were analyzed. We do not know exactly whether, for example, physical distance between the newcomer and his or her leader might affect the interaction in certain work group structures. There are many intriguing areas for future research.
Acknowledgment
We thank this journal’s four anonymous reviewers of our paper for the thoroughness of the feedback and direction they provided. This research is supported by the Faculty Research Grants of Macau University of Science and Technology (0380) and Macau Foundation Grants (0398h).
