Multilevel Model: TFL, Skill Development, Knowledge & Creativity

This literature review examines the impact of dual-focused transformational leadership (TFL) on employee creativity, addressing the challenge of fostering both individual and team creativity. It introduces a multilevel model connecting dual-focused TFL to creativity, incorporating skill development at the individual level and knowledge sharing at the team level as mediating mechanisms. The review differentiates between individual-focused TFL, which emphasizes individual skill development, and team-focused TFL, which promotes team knowledge sharing. Findings from high-technology firms indicate that individual-focused TFL positively influences individual creativity through skill development, while team-focused TFL impacts team creativity through knowledge sharing. Furthermore, knowledge sharing moderates the relationship between individual-focused TFL, skill development, and individual creativity, highlighting the synergistic effects of individual and team dynamics. The study contributes to the understanding of how leaders can facilitate multilevel creativity by examining the differential impacts of dual-focused TFL and identifying level-specific mechanisms that connect TFL to creativity.

See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/308293733
Enhancing employee creativity via individual skill development and team
knowledge sharing: Influences of dual‐focused transformational leadership
Article in Journal of Organizational Behavior · September 2016
DOI: 10.1002/job.2134
CITATIONS
51
READS
5,749
4 authors, including:
Some of the authors of this publication are also working on these related projects:
Examining Multilevel Factors' Influences on Organizational Creativity and Innovation and Its Dynamic Mechanism in Enterprise TransformationView project
Gerardo R UngsonView project
Yuntao Dong
University of Connecticut
5 PUBLICATIONS73CITATIONS
SEE PROFILE
Zhi-Xue Zhang
Peking University
66PUBLICATIONS1,438CITATIONS
SEE PROFILE
Chenwei Li
San Francisco State University
17PUBLICATIONS188CITATIONS
SEE PROFILE

Enhancing employee creativity via individual ski
development and team knowledge sharing:
Influences of dual-focused transformational
leadership
YUNTAO DONG1*, KATHRYN M. BARTOL 2, ZHI-XUE ZHANG 3 AND CHENWEI LI 4†
1University of Connecticut, Storrs, Connecticut, U.S.A.
2University of Maryland, College Park, College Park, Maryland, U.S.A.
3Peking University, Beijing, China
4San Francisco State University, San Francisco, California, U.S.A.
Summary Addressing the challenges faced by team leaders in fostering both individual and team creativity
developed and tested a multilevelmodelconnecting dual-focused transformationalleadership (TFL)and
creativity and incorporating intervening mechanisms at the two levels. Using multilevel, multiso
data from individual members, team leaders, and direct supervisors in high-technology firms, we
individual-focused TFL had a positive indirecteffecton individualcreativity via individualskill develop-
ment, whereas team-focused TFL impacted team creativity partially through its influence on tea
sharing. We also found that knowledge sharing constituted a cross-level contextual factor that m
relationship among individual-focused TFL,skill development,and individualcreativity.We discuss the
theoreticaland practicalimplications of this research and offer suggestions for future research.Copyright
© 2016 John Wiley & Sons, Ltd.
Keywords: transformationalleadership;employeecreativity;skill development;knowledgesharing;
multilevel
Employee creativity, defined as the generation of novel and useful ideas (Amabile, 1988; Zhou & Sha
critical to organizational survival and effectiveness. There has been increasing research interest in ex
leaders might do to encourage the production of creative outcomes (Anderson, Potočnik, & Zhou, 201
of teams,in particular,this can presenta specialchallenge.On the one hand,ideas are ultimately offered up by
individuals,and hence,it is useful for leaders to develop individuals'knowledge and skills needed for creativity.
On the other hand,related research suggests that team creativity is more than the sum of its individual
requires the exchange of knowledge among team members. Emphasizing promoting individuals'development (Dvir,
Eden,Avolio, & Shamir,2002)while encouraging collective contribution (Eisenbeiss,Van Knippenberg,&
Boerner, 2008), transformational leadership (TFL) is particularly well suited to set in motion appropria
at both the individual and team levels to handle this dual challenge.
Yet, in a recent meta-analysis, Rosing, Frese, and Bausch (2011) have pointed to inconsistencies in
on TFL–creativity relationship and suggested that the high degree of variation found in the relationsh
to the lack of clarification of the levels of analyses. In fact, recent theoretical advancements regardin
effective transformational leaders have different emphases when managing individuals and teams (L
Boyle, 2016; Li, Shang, Liu, & Xi, 2014; Wang & Howell, 2010; Wu, Tsui, & Kinicki, 2010), with some b
*Correspondence to: Yuntao Dong, University of Connecticut, 2100 Hillside Road, Unit 1041, Storrs, CT 06269, U.S.A. E-mail: y
[e-mail]
†This research was initiated when Professor Chenwei Li was at Indiana University - Purdue University Fort Wayne, Fort Wayne,
Copyright © 2016 John Wiley & Sons, Ltd.
Received 28 May 2015
Revised 30 July 2016, Accepted 10 August 2016
Journal of Organizational Behavior, J. Organiz. Behav. (2016)
Published online in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/job.2134
Research Article

most meaningfully targeted at individuals (individual-focused TFL) and other behaviors more properly dir
the team (team-focused TFL). Thus, the dual-focused conceptualization of TFL may have the potential to
creativity research. Li et al. (2016) have made a first attempt to adopt dual-focused TFL in this realm by
predictinnovation in teams.However,there is widespread recognition thatinnovation encompasses both idea
generation (creativity)and implementation,which are increasingly viewed as involving substantially different
dynamics and mechanisms (Anderson etal.,2014;Rosing etal.,2011).Hence,in this study,we focus squarely
on multilevel associations between dual-focused TFL and creativity.
Although TFL is recognized as an influential enabler for creativity, there have been diverse and mixed
what pathways a leader may take to affect individual and team creativity. One major line of research has
the role of the leader as a facilitator of follower creativity (Mainemelis, Kark, & Epitropaki, 2015). Here, w
thatresearch stream by exploring the possibility thatdual-focused TFL increases the prospectthatthe followers
themselves can step up to the creativity requirements of the job,while also stimulating the team toward greater
creativity production.Specifically,individual-focused TFL tends to “stimulate individuals to develop their own
skills” (Li et al., 2016, p. 4), while team-focused TFL motivates team members to offer information out of
to benefitthe collective (Zhang,Tsui, & Wang,2011)and enables “transmission and sharing ofknowledge”
(García-Morales, Lloréns-Montes, & Verdú-Jover, 2008, p. 300). Enhanced individual skills and knowledge
may, in turn, help the individual and the group, respectively, enact their creative potential. Hence, indivi
development and team knowledge sharing may be important mechanisms linking dual-focused TFL and
However, research has yet to examine these mechanisms simultaneously under a dual-focused framew
tions remain regarding how dual-focused TFL may uniquely and synergistically impact individual and tea
ity. This is important because, as Gong, Kim, Lee, and Zhu (2013, p. 844) have noted, “individuals must
back into the study ofteam creativity” as team creativity depends on the foundationalindividualcapability to
generate ideas. At the same time, as Gilson, Lim, Luciano, and Choi (2013) have claimed, individual deve
and creativity are constantly shaped by team knowledge exchange. These arguments highlight the pote
tages of integrating dual-focused TFL and creativity and the associated mechanisms in a multilevel fram
Accordingly,we propose and testa theoreticalmodelthatexamines how dual-focused TFL may influence
individualand team creativity via separate channels.Integrating the dual-focused TFL perspective and creativity
literature,we consider skilldevelopmentatthe individualleveland knowledge sharing atthe team levelas the
mediating mechanisms. Further, in view of calls for understanding how the multilevel mediators of TFL m
with each other (Van Knippenberg & Sitkin,2013),we investigate the cross-levelinfluence of team knowledge
sharing on the individual-level relationships.
We aim to make three significant contributions. First, we examine the differential impacts of individual
team-focused TFL on creativity, demonstrating the utility of dual-focused TFL in the creativity realm. Sec
tifying individual skill development and team knowledge sharing as influence channels of individual-focu
team-focused TFL, our study addresses the need for research on level-specific mechanisms that connect
ity (Van Knippenberg & Sitkin, 2013) and contributes to the broader literature on creative leadership wit
differentmeans through which a leader may facilitate multilevelcreativity (Mainemelis etal.,2015).Third,by
simultaneously examining the multilevel mediators, we are able to provide insights regarding the cross-l
on individual creativity while bringing a conceptual and empirical integration to “findings related to the d
effects of team dynamics and individual contributions” on employee creativity (Li et al., 2016, p. 2).
Theory and Hypotheses
In this section, we trace the development of our model by first explicating a dual-focused conceptualizat
We next examine the mediating relationships explaining how individual-focused TFL boosts employee cr
enhancing individualskill developmentand how team-focused TFL fosters team creativity via promoting team
Y. DONG ET AL.
Copyright © 2016 John Wiley & Sons, Ltd. J. Organiz. Behav. (2016)
DOI: 10.1002/job

members'knowledge sharing. Finally, we discuss the cross-level moderating effect of team knowledge
individual creativity. The hypothesized model is depicted in Figure 1.
Dual-focused transformational leadership
Historically, the conceptualization of TFL has not been clear about the levels of analyses and research
TFL by using mixed referents (Van Knippenberg & Sitkin, 2013). Yet, a number of researchers have no
sibility that TFL could operate at both individual and team levels.For example,Kark and Shamir (2002) viewed
TFL as a multifaceted and complex form of leadership that may function at multiple levels. Wu et al. (
that TFL behaviors could have either an individual focus that addresses the needs of each team mem
focus thatinfluences a team as a whole.Consistentwith these viewpoints,Wang and Howell(2010) proposed a
dual-focused modelof TFL and identified specific leader behaviors thatare better suited for consideration atthe
individual level or at the team level.
Drawing on related priorresearch,we pointoutthatthe foundationallogic underlying the differentiation of
individual-focused and team-focused TFL is thatmotivating individuals and teams requires differentemphases
and varied behaviorsfrom the leader.Specifically,following Podsakoff,MacKenzie,Moorman,and Fetter
(1990), Wu et al. (2010), and Wang and Howell's (2010) work, we conceptualize individual-focused TF
emphasizing the uniqueness of each individual member by paying close attention to individuals'needs, intellectually
challenging them, and expressing high expectations for personal development and excellence. As Wu
suggested,individual-focused leaders“considersubordinates'unique capabilitiesand characteristics” because
“abilities and intelligence vary among individuals” (p.93).Individualized consideration,intellectualstimulation,
and performance expectations may be customized fordifferentindividuals and may notbe shared by allteam
members.Hence,these behaviors focus more on individuals than on collective interests and are mainly
at individuals.Consequently,they tend to shape individualoutcomesin ways thatare germane to specific
employees.
On the other hand,we define team-focused TFL as leaders emphasizing common grounds,shared values,and
ideology among team members by articulating a compelling vision,constituting a role modelfor the team,and
fostering acceptance of collective goals.These behaviors are more appropriately conceptualized at the team
than at the individual level because they tone down individual differences and serve as “ambient stim
followers to prioritize collective interests (Chen, Farh, Campbell, Wu, & Wu, 2013). As a result, team-f
tends to exert influence on a group as a whole. It is worth noting that the differentiation between ind
and team-focused TFL does not mean that one set of behaviors can only occur at its designated levelthe
Figure 1.Theoretical model of dual-focused TFL and individual and team creativity
INFLUENCES OF DUAL-FOCUSED TFL ON CREATIVITY
Copyright © 2016 John Wiley & Sons, Ltd. J. Organiz. Behav. (2016)
DOI: 10.1002/job

conceptualization points out that some behaviors are conceptually more functional and more likely to op
certain level than at the other level.
Impacts of individual-focused TFL on creativity via individual skill development
Creativity theories suggest that effective leadership can serve as an enabler that helps individuals acqui
knowledge and skills, which are a key component for creating new ideas or products (Amabile, 1988; An
2014). This view is supported by studies from related fields. For example, in considering creativity as a s
of proactivity, Parker and Wu (2014, p. 393) proposed that “leaders can influence the proactivity of their
through fostering the development of knowledge, skills, and abilities.” Further, the developmental notion
individual-focused TFL, given its emphasis on mentoring, coaching, and challenging individual followers b
their needs and abilities.As such,follower developmentconstitutes a principalinfluence process through which
individual-focused TFL may foster individual creativity. We define individual skill development as one's a
ment in developmental activities in order to obtain work-related knowledge and skills to facilitate long-te
acquisition and enhancement related to one's work (Noe, 1996; Pulakos, Arad, Donovan, & Plamondon, 2
Relevant evidence suggests that individual-focused TFL can influence follower development by commu
high expectations for excellence and superior performance to stimulate followers'intrinsic needs for growth (Bass &
Avolio, 1994; Wang & Howell, 2010). Consequently, followers will more actively participate in developme
ities. For example, Dvir et al. (2002) found that transformational leaders motivate followers to exert extr
satisfy their own self-actualization needs. In addition, through intellectual stimulation, transformational l
encourage followers to challenge traditions and to think about problems from different perspectives,all of which
require followers to master new skills and to shift viewpoints via continuous development (Shin & Zhou,2003).
Finally,transformationalleaders emphasize individualized consideration,provide support,and show respectfor
individuals'needs for development.Along these lines,Maurer and colleagues (Maurer & Tarulli,1994;Maurer,
Weiss,& Barbeite,2003) found thatsupervisor supportfor developmentwas positively related to subordinates'
intention to participate in development activities as well as subsequent actual participation (e.g., attend
reading journals, or initiating new projects). Taken together, we expect individual-focused TFL to motivat
ual followers to actively engage in skill development.
Hypothesis 1: Individual-focused TFL is positively related to individual skill development.
Creativity scholars(e.g.,Amabile,1988)have posited thatthe extentto which individualsdevelop their
knowledge and skills in a given area of interest is one of “the major influences on output of creative idea
and is “necessary for … creativity to be produced” (p. 137). These skills comprise fundamental resources
uals to put existing information and newly generated ideas together in novel combinations (Anderson et
other words, factual knowledge, technical skills, and work experience in the domain of interest allow an
understand complexities and generate a set of response possibilities from which new responses can be s
larger the set of possibilities, the more numerous the alternatives available for producing something new
individuals who engage in skill development are more likely to have an increased range and depth of kno
more developed skills, providing them the basis for critical thinking that is needed to creatively solve pro
improve current processes (Zhang & Bartol, 2010). Woodman, Sawyer, and Griffin (1993, p. 301) quoted
insight: “Invention is little more than a new combination of those images, which have been previously ga
deposited in the memory.Nothing can be made ofnothing.He who has laid up no materialcan produce no
combination” (quoted in Offner, 1990).
Supporting these arguments, Chen, Shih, and Yeh (2011) showed that individual possession of differen
was critical for public sector employees to connect different disciplines of knowledge and engage in the
thinking thatis crucialfor creativity.Similarly,Yang,Lee,and Cheng (2016) provided some relevantevidence
Y. DONG ET AL.
Copyright © 2016 John Wiley & Sons, Ltd. J. Organiz. Behav. (2016)
DOI: 10.1002/job

by showing that frontline bank tellers who possessed more knowledge and skills were likely to engag
thinking and identify new opportunities in the service realm. Finally, Choi (2004) found that creative a
refers to skills or competencies relevant to creative performance, such as the ability to look at proble
perspectives,was significantly related to individualcreativity.Taken together,ourtheorizing and the evidence
suggest that individual-focused TFL may enhance follower creativity because it is associated with kno
skill acquisition, a major predictor of individual creative performance.
Hypothesis2: Individualskill developmentmediatesthe relationship between individual-focused TFL and
individual creativity.
Impacts of team-focused TFL on team creativity as mediated by knowledge sharing
Enhancing team creativity requires leaders not only to develop creative individuals but also to promo
communication and information exchange (Hargadon & Bechky, 2006). As Taggar (2002) has pointed
should reach a higher level of creativity when they contain both creative members and effective proc
members can collectively approach and utilize knowledge available within the team. Due to its emph
vision and collective goals,team-focused TFL may be particularly effective in involving team membersin
knowledge sharing activities, which, in turn, help the team enact its creative potential. We define tea
sharing as the extent to which team members share task-relevant ideas, information, and suggestion
(Srivastava, Bartol, & Locke, 2006).
By communicating and promoting collective vision and values, team-focused transformational lead
membersunderstand thatindividualinputsof usefulknowledge and information are valuable to team goal
accomplishment.As a result,team members are likely to feel more inclined to engage in such efforts as off
constructive suggestions and sharing unique information so as to contribute to the achievementof mutualgoals
(Eisenbeiss etal.,2008).Moreover,by providing an appropriate role modeland facilitating team acceptance of
collective objectives, team-focused TFL may shape a deeper understanding and appreciation of contr
all members (Li et al., 2014; Shin, Kim, Lee, & Bian, 2012). As team members feel that their inputs ar
are willing to take the opportunity to share their knowledge (Detert& Burris,2007).In sampling research and
development(R&D) teams in a multinationalpharmaceuticalcompany,Kearney and Gebert(2009) found that
TFL was effective in fostering the exchange and elaboration of task-relevant information when the te
of specialized members.Consistently,Li etal. (2014) found thatgroup-focused leadership increased individual
members'knowledge-providing and knowledge-collecting behaviors. Based on these arguments, we hy
positive impact of team-focused TFL on team knowledge sharing.
Hypothesis 3: Team-focused TFL is positively related to team knowledge sharing.
The positive impactof team knowledge sharing on team creativity is consistentwith the suggestion thatthe
communication of individual knowledge in a team is a viable resource for the team to generate new i
et al.,2013; Zhang et al.,2011).In line with previous research (e.g.,Gino,Argote,Miron-Spektor,& Todorova,
2010;Hülsheger,Anderson,& Salgado,2009),we refer to team creativity as the combination of newness an
usefulness of ideas that are developed by the team. Team creativity is not simply the aggregation of
by individual members; rather, it involves team members collectively processing information, conside
views, and eventually producing creative outcomes.
Specifically,because exchange of diverse information helps in boosting the repository of available ex
skills, and knowledge in the team, it enables the team to utilize and integrate the resources to accom
tasks, such as those of developing new products or procedures (Gardner, Gino, & Staats, 2012). For e
and Choi(2012)argued thata greaterteam knowledge stock offers more opportunities to recombine existi
INFLUENCES OF DUAL-FOCUSED TFL ON CREATIVITY
Copyright © 2016 John Wiley & Sons, Ltd. J. Organiz. Behav. (2016)
DOI: 10.1002/job

information and ideas by making rich cognitive resources and diverse approaches available. These schol
evidence that if teams can identify relevant knowledge and activate the value of knowledge distributed a
bers, they will achieve higher levels of creativity. In contrast, if individual experience and resources are n
the cognitive resources within a team remain underutilized (Gino et al., 2010; Griffith & Sawyer, 2010). I
knowledge sharing also leads to a more comprehensive consideration of information, which is a necessa
enabling collective creativity (Gong et al., 2013).
The positive association between knowledge sharing and creativity has recently received some empiric
in a variety of teams,including research and developmentteams,manufacturing groups,and managementteams
(Gong etal.,2013;Sung & Choi,2012;Zhang etal.,2011).Based on a meta-analysis,Hülsheger etal. (2009)
concluded that internalcommunication of knowledge and other teamwork-related subjects among team mem
emerged as one of the most powerful agents of new idea generation and implementation in the team.Combining
with arguments underlying Hypothesis 3,we suggestthatteam-focused TFL willbe positively related to team
knowledge sharing, which will then lead to higher levels of team creativity.
Hypothesis 4: Team knowledge sharing mediates the relationship between team-focused TFL and team
Cross-level moderating impact by team knowledge sharing
According to Amabile (1988),open communication systems for knowledge exchange among differentorganiza-
tionalmembers may influence the extentto which domain skills fosterindividualcreativity.Therefore,team
knowledge sharing,as an indicator of open communication in work groups,may have a cross-levelimpacton
the relationship between individual skill development and creativity.Specifically,when sharing knowledge,team
members are exposed to various points of view and multiple alternatives,which may mutually inspire individual
team members (Hirst,Van Knippenberg,& Zhou,2009;Homan,Van Knippenberg,Van Kleef,& De Dreu,
2007).Individualmembers'knowledge poolmay be enlarged and theirdivergentthinking may be activated
and enhanced (Sheremata,2000).In other words,the sharing provides team members with more valuable infor-
mation and the discussion of shared knowledge can inspire them to develop new insights and strategies
ing problems.Team members can build on others'contributions to supplement their existing resources and aid in
producing their own creative alternatives. Hence, whereas one's skills and knowledge comprise the basic
blocks for individualcreativity,team knowledge sharing helps individualmembers to better utilize their existing
knowledge in generating novelideas (Zhou,Shin,Brass,Choi,& Zhang,2009).In line with this logic,Gong
etal. (2013) showed thatthrough exchanges with other team members and being exposed to diverse ideas,an
individual team member may enhance his or her divergent thinking that is conducive to creativity.They also ar-
gued that knowledge sharing may be particularly important for R&D teams because those team member
rely on high-quality information exchange to tackle complex problems and develop new products and se
a regular base.
Hypothesis 5a: Team knowledge sharing moderates the relationship between individualskilldevelopmentand
individual creativity, such that the relationship is stronger when there are higher rather than lower lev
knowledge sharing.
Considering also the individual-level relationships among TFL, skill development, and creativity hypoth
Hypothesis 2, we further propose that team knowledge sharing moderates the effect of individual-focuse
creativity via developing individual skills.
Hypothesis 5b: Team knowledge sharing moderates the mediated relationship between individual-focu
and individual creativity via skill development, such that the relationship is stronger when there are hi
than lower levels of team knowledge sharing.
Y. DONG ET AL.
Copyright © 2016 John Wiley & Sons, Ltd. J. Organiz. Behav. (2016)
DOI: 10.1002/job

Method
Research site and participants
We collected data from eightcompanies in a major high-technology developmentzone located in northwestern
China.We sampled their R&D teams which consisted of professional-level employees working interdep
such as software engineers and new product developers.One of the authors visited these companies to distribut
and collect questionnaires during their working hours. Data were collected from team members, team
supervisors to whom the team directly reports. In the member survey, team members offered demog
tion and assessed the team leader's behaviors,team knowledge sharing,and task interdependence.In the leader
survey,team leaders evaluated each member's skilldevelopmentbehaviors and creativity.In a separate survey,
supervisors rated each team's creativity. All surveys were distributed to the participants in separate e
completed, the survey was put back in the envelope and then collected by the researcher. As all part
questions on theirown survey and theirresponses were keptconfidential,we reduced participants'concern in
answering the questions,especially those requiring team members to assess their team leader.The finalsample
consisted of 171 individuals in 43 teams from eight companies.Among the team members,the average age was
29 years, the average team tenure was 24 months, and 60 percent were male.
Measures
All surveys were translated from English into Chinese and then back translated into English to ensure
meaning.Unless otherwise noted,all the variables were measured on 7-point Likert-type scales (1 = Strong
agree, 7 = Strongly agree). As shown in Table 1, all scales exhibited adequate reliability.
Table 1.aDescriptive statistics, correlations, and reliabilities.
Individual-level variables Mean SD 1 2 3 4 5 6
1. Age 29.01 5.29
2. Gender (1 = male, 2 = female)1.39 0.49 .12
3. Team tenure (months) 24.43 25.91 .34** .23**
4. Team-focused TFL 5.80 0.85 .09 -.16* .14+ (.83)
5. Individual-focused TFL 5.48 0.92 .35** .29** .19* .67*** (.94)
6. Individual skill development5.60 0.75 .17* .08 .12 .02 .16* (.85)
7. Individual creativity 5.05 0.95 .10 .02 .13 .05 .10 .50*** (.87)
Team-level variables Mean SD 1 2 3 4 5
1. Team size 7.30 3.81
2. Team task interdependence6.04 0.33 .06 (.73)
3. Individual-focused TFL (agg.)5.54 0.64 .07 .27+
4. Team-focused TFL (agg.) 5.75 0.62 .24 .34* .74***
5. Team knowledge sharing 5.55 0.55 .18 .11 .50*** .38* (.92)
6. Team creativity 3.95 0.85 .09 .29+ .34* .32* .42* (.92)
an = 171 for individual-level data and n = 43 for team-level data. agg. = aggregation. Reliabilities for the scales are in parent
along the diagonal.
***p < .001; **p < .01; *p < .05;+
p < .10.
INFLUENCES OF DUAL-FOCUSED TFL ON CREATIVITY
Copyright © 2016 John Wiley & Sons, Ltd. J. Organiz. Behav. (2016)
DOI: 10.1002/job

Dual-focused transformational leadership
Transformational leadership was assessed using the 14-item version of Podsakoff et al.'s (1990) measure
also been used by Kirkman, Chen, Farh, Chen, and Lowe (2009). Subsequent to the initiation of our data
we became aware of an emerging dual-focused TFL scale (Wang & Howell, 2010), which was based on th
TFL measures (e.g.,Avolio & Bass,2004;Podsakoff etal.,1990).Accordingly,we conducted a supplementary
study thatconfirmed the measurementequivalence between the scale we used and Wang and Howell's scale
(Appendix A).Podsakoff etal. (1990) identified six classes of TFL behaviors.Consistentwith the dual-focused
TFL conceptualization,offering individualized support,providing intellectualstimulations,and expressing high
performance expectations constituted individual-focused TFL and articulating a team vision,providing an appro-
priate model,and fostering acceptance of team goals constituted team-focused TFL.The referents in our items
reflected the appropriate targetof influence (i.e.,individualmembers or the team).Sample items for individual-
focused TFL included [The leader] “shows respectfor my personalfeelings” and “has stimulated me to rethink
the way Ido things.” Sample items forteam-focused TFL included [The leader]“articulates a vision forthe
team” and “facilitates the acceptance of group goals.”
In keeping with previous research (e.g.,Li etal.,2016;Wang & Howell,2010),we averaged allitems atthe
individual level to create an overall score for the individual-focused TFL scale.To obtain the team-focused TFL,
we averaged allteam-focused items for each individualand then aggregated members'team-focused TFL scores
for each team.We computed rwg(j)to assess interrater agreement. The average rwg(j)across the 43 teams was .97,
indicating a high level of within-team agreement. We also obtained the intraclass correlation (ICC1) and
of group mean (ICC2). The ICC1 and ICC2 values were .23 (p < .001) and .54, respectively. Taking all the
into account, we concluded that aggregation of team member ratings of team-focused TFL was warrante
Individual skill development
Individual skill development was reported by team leaders using a six-item measure based on the work b
et al. (2000) and London and Mone (1987). It captured an individual team member's skill development b
the work context that were observed by the team leader. A sample item was “This team member works
his/her knowledge and skills up-to-date so he/she can work effectively.” We averaged all items to obtain
scale.
Individual creativity
We measured individualcreativity using the well-established four-item scale reported by Farmer,Tierney,and
Kung-McIntyre (2003). A sample item was “This team member seeks new ideas and ways to solve proble
Team knowledge sharing
Team knowledge sharing was assessed by the six team knowledge sharing items developed by Bartol, Li
and Wu (2009).A sample item was “There is a lot of exchange of information,knowledge,and sharing of skills
among members in our team.” The average rwg(j)across the 43 teams was .98. The ICC1 value was .17 (p < .01),
and the ICC2 value was .44. We aggregated the individual-level knowledge sharing scores for each team
the respective team-level construct.
Team creativity
Supervisors were asked to rate each team's creativity by using a four-item scale developed by Shin and
on a 7-pointscale ranging from 1 (Needs much improvement)to 7 (Excellent).Supervisors were provided the
definition of team creativity. A sample item was “How creative do you consider this team to be?”
Covariates
At the individual level, we controlled for individuals'age, gender, and team tenure, which have been found to impa
individual learning and creativity (e.g., Gong et al., 2013). At the team level, we controlled for team size
Y. DONG ET AL.
Copyright © 2016 John Wiley & Sons, Ltd. J. Organiz. Behav. (2016)
DOI: 10.1002/job

team leaders to reportthe totalnumber of team members.Team size has been suggested to influence employee
creativity and various team processes (e.g., Eisenbeiss et al., 2008; Hirst et al., 2009). We also contro
task interdependence because itmay have a distinctimpacton team knowledge sharing and employee creativity
(Gong et al.,2013; Li et al.,2016).Task interdependence was assessed using the four-item scale from Kirk
Rosen, Tesluk, and Gibson (2004) and aggregated to the team level.
Confirmatory factor analysis
We conducted a series of confirmatory factor analyses (CFAs) to examine the distinctiveness of the k
the study. The overall model fit was assessed by the comparative fit index (CFI), the Tucker–Lewis ind
standardized root mean square residual (SRMR), and the root mean square error of approximation (R
hypothesized five-factormodel(i.e.,individual-focused TFL,team-focused TFL,skill development,individual
creativity,and knowledgesharing)indicated agood fit to the data(χ2(383)= 690.53,CFI = .92,TLI = .90,
SRMR = .07, RMSEA = .06). All indicators loaded significantly (p < .05) onto the intended latent varia
parison with a four-factor model, in which individual-focused TFL and team-focused TFL were loaded
(χ2(387)= 776.87, CFI = .89, TLI = .88, SRMR = .08, RMSEA = .06), showed that the hypothesized mode
significantly better than the alternative model(Δχ2(4)= 86.34,p < .001).We also compared the five-factor model
with another four-factor model,in which team leader-rated individual skill development and individual crea
were loaded on one factor. The model did not fit the data well (χ2(387)= 782.56, CFI = .89, TLI = .88, SRMR = .08,
RMSEA = .07) and was significant worse than the hypothesized model (Δχ2(4)= 92.03, p < .001). Finally, we found
that compared with a three-factor model, in which all variables reported by team members (individua
team-focused TFL,and knowledge sharing)were loaded on one factor(χ2(390)= 1211.78,CFI = .77,TLI = .75,
SRMR = .11,RMSEA = .09),the hypothesized models fitthe data significantly better (Δχ2(7)= 521.25,p < .001).
These results supported the discriminant validity of the measures used in the study.
Analytical approach
We used hierarchicallinear modeling (HLM) to testthe multilevelmodel.We included an intercept-only model
atthe organization levelto controlfor organizationaleffects.Hence,for the individual-leveland cross-levelre-
lationships,we applied HLM3 with team members specified atLevel1, teams atLevel2, and organizations at
Level 3.For the team-level relationships,we conducted HLM2 with teams at Level 1 and organizations at Lev
2. Following the recommendations by Zhang,Zyphur,and Preacher (2009),we group-mean centered individual
skill development and added its group mean back to the Level 2 intercept-only model as covariates f
levelinteractions.When testing the effectof individual-focused TFL,we included team-focused TFL to control
for its influence,whereaswhen testing the effectof team-focused TFL,we controlled forthe influence of
individual-focused TFL.This strategy allows us to demonstrate the distincteffects ofone aspectof TFL at a
given levelaftercontrolling forthe effectsof the otherTFL aspectat a differentlevel.To examine the
significance of the mediations and the moderated mediated relationships,we employed a Monte Carlo simulation
procedure using the open-source software R (Preacher & Selig,2012).This method more accurately reflects the
asymmetric nature of the sampling distribution of a mediating effect.
Results
Table 1 presents descriptive statistics and bivariate correlations for all variables at both individual an
INFLUENCES OF DUAL-FOCUSED TFL ON CREATIVITY
Copyright © 2016 John Wiley & Sons, Ltd. J. Organiz. Behav. (2016)
DOI: 10.1002/job

Results of HLM null models
Prior to hypothesis testing,we examined whether there was significantbetween-team and between-organization
variance in the outcome variables by running null (intercept-only) models.The results showed that 29.33 percent
of the variance in individualcreativity resided atthe team level(χ2
(35)= 85.47,p < .001) and 0.31 percentof the
variance resided at the organization level (χ2
(7)= 10.50, p > .05). In addition, 14.97 percent of the variance in indiv
ual development resided at the team level (χ2
(35)= 58.42; p < .01) and 8.78 percent of the variance resided at the
organization level (χ2
(7)= 17.53; p < .05). We also found that 19.68 percent of the variance in team creativity
at the organization level(χ2
(7)= 15.86,p < .05).The results supported using HLM3 forindividual-levelmodel
estimation and HLM2 for team-level model estimation.
Hypothesis testing
Individual-level relationships
The three-level HLM results are summarized in Table 2. Hypothesis 1 stated that individual-focused TFL i
tively related to skilldevelopment.As shown in Model1 of Table 2,individual-focused TFL was significantly
related to individual skill development (γ = .21, p < .05). Hypothesis 1 was supported
Hypothesis 2 proposed that skill development mediates the relationship between individual-focused TF
ativity.As shown in Table 2,although individualTFL did nothave a significantimpacton individualcreativity
(γ = .08,p > .05;Model2), individualskill developmentand creativity were positively related when individual
TFL was also included in the model(γ = .61,p < .001;Model3), indicating an indirectrelationship between
Table 2.aResults of hierarchical linear modeling for individual-level and cross-level relationships.
DV: Individual DV: Individual
Skill Development Creativity
Variables Model 1 Model 2 Model 3 Model 4
Level 1 and Level 2 covariates
Age .01 .01 .00 .01
Gender .00 .16 .16 .18
Team tenure .00 .00 .00 .00
Team-focused TFL .11 .05 .02 .02
Team size .00 .02 .02 .02
Team task interdependence .15 .26 .28 .29
Team knowledge sharing .22 .21 .34 .37*
Team mean of .67***
Individual skill development
Level 1 predictors
Individual-focused TFL .21* .08 .05 .04
Individual skill development .61*** .60***
Cross-level interaction
Individual skill development .40*
*Team knowledge sharing
Pseudo R2 .06 .04 .29 .30
an = 171 for individual-level data (Level 1), n = 43 for team-level data (Level 2), and n = 8 for organization-level data (Level 3). Th
organization-level independent variables. DV = dependent variable. Pseudo R2 indicates the amount of total variance in the dependent variable
explained by predictors in the model.
***p < .001; *p < .05.
Y. DONG ET AL.
Copyright © 2016 John Wiley & Sons, Ltd. J. Organiz. Behav. (2016)
DOI: 10.1002/job

individual-focused TFL and individual creativity via skill development.1 We further examined the significance of
the indirecteffectusing the Monte Carlo simulation procedure with 20,000 replications.Results reported in
Table 4 showed that the indirect effect was 0.13, with a 95 percent confidence interval of [0.025, 0.2
Hypothesis 2 received partial support.
Team-level relationships
Hypotheses 3 and 4 were tested in two-level HLM, and the results are presented in Table 3. Hypothes
team-focused TFL is positively related to team knowledge sharing.As shown in Model1 of Table 3,after
controlling for the effect of aggregated individual-focused TFL, team-focused TFL was significantly rel
knowledge sharing (γ = .17, p < .05), supporting Hypothesis 3.
Hypothesis 4 proposed that team knowledge sharing mediates the relationship between team-focu
team creativity. As shown in Models 2 and 3 of Table 3, there was a significant relationship between t
TFL and team creativity (γ = .50, p < .05), as well as a significant relationship between team knowled
team creativity when team TFL was also included (γ = .42, p < .01), indicating that team knowledge
mediated the relationship between team-focused TFL and team creativity. Results from the Monte Ca
revealed thatthe indirecteffectwas 0.07,with a 95 percentconfidence intervalof [0.011,0.160].The analyses
supported Hypothesis 4.
Cross-level interaction
Hypothesis5a proposed thatteam knowledgesharing strengthensthe relationship between individualskill
developmentand individualcreativity.As shown in Model4 of Table 2,the cross-levelmoderation ofteam
knowledgesharingand individualskill developmentwas significantin predictingindividualcreativity
(γ = .40, p < .05). Using the multilevel interaction computational tool developed by Preacher, Curran
(2006),we plotted this interaction atone standard deviation above and below the mean ofteam knowledge
sharing.The plotin Figure 2 and the simple slope tests suggested thatindividuals'developmentwas positively
associated with their creativity when there were higher levels of team knowledge sharing (b = .38, p
positive relationship was stronger when there were lower levels of team knowledge sharing (b = .83,
the predicted interaction was significant, the pattern of the interacting effect was not as expected. H
not supported.
Hypothesis 5b predicted that team knowledge sharing moderates the relationship between individu
and individual creativity via skill development. The conditional indirect effects and the 95 percent con
vals estimated using the Monte Carlo simulation procedure are presented in Table 4. Specifically, we
indirect effect was positive and significant when team knowledge sharing was high (estimate = .08, C95%= [0.010,
0.184]),and itwas even stronger when team knowledge sharing was low (estimate = .18,CI95%= [.035,.341]).
These findings suggested a conditionalindirectrelationship,yetthe pattern was inconsistentwith our prediction.
Hence, Hypothesis 5b was not supported.
Discussion
This research delineates the multilevel influences of dual-focused TFL on employee creativity. Analys
that individual-focused TFL had an indirect effect on individual creativity via individual skill developm
team knowledge sharing partially mediated the relationship between team-focused TFL and team cre
1We followed Mathieu and Taylor (2006) and Hayes's (2009) suggestion to differentiate a mediation effect,which requires the present of the
X ➜ Y relationship,from an indirect effect,which does notrely on the existence of the X ➜ Y relationship.Hence,we refer the relationship
among individual-focused TFL, individual skill development, and creativity found in the study as an indirect effect in the rest o
INFLUENCES OF DUAL-FOCUSED TFL ON CREATIVITY
Copyright © 2016 John Wiley & Sons, Ltd. J. Organiz. Behav. (2016)
DOI: 10.1002/job