Fluent Essays

Business and Management Topic: Transformational Leadership Theory Behaviors Type of work: Essay Level: Undergraduate Number of pages: 3 pages = 250/- Grade: High Quality (Normal Charge) Formatting style: APA Language Style: English (U.S.) Sources: 3 Website Region: United States e Annotated bibliography of the 3 provided articles. Provide a closing paragraph that draws the articles together with these common themes communication, motivation, influence, and stimulation. Pak. J. Commer. Soc. Sci. 2012 Vol. 6 (2), 357-371 A Study of Relationship between Leader Behaviors and Subordinate Job Expectancies: A Path-Goal Approach Sikandar Hayyat Malik Senior Joint Director, State Bank of Pakistan, Karachi E-mail: [email protected] E-mail: [email protected] Abstract This study investigates relationship between leader behavior (directive, supportive, participative and achievement-oriented) of corporate managers and subordinates’ job expectancies using House (1974) Path-goal model of leadership. The results reveal that leader behavior affects subordinates’ job expectancies. The situational factors (task structure, role ambiguity, stress, need for autonomy, locus of control, need for achievement and perception about abilities) affect subordinates’ job expectancies (I&II). While subordinates’ attributes/characteristics (age, gender, qualification, rank, experience and length of service under the current supervisor) do not affect job expectancies (I&II) except for rank/position and expectancy-II. Path goal theory predicts that directive leader behavior will be more effective for the subordinates with high need for achievement because directive leader through clarifying path guides subordinates. Similarly, participative leader behavior is also effective as he consults with subordinates in setting, clarifying and achieving goals. The results of this study reveal that there is an inverse relationship between subordinates’ job expectancy (I&II). According to Yukl (2006), for subordinates with high need for autonomy, participative leader behavior will increase the intrinsic valence of work Keywords: Leadership, Subordinates’ job expectancies, Path-goal theory, Leader behavior. 1. Introduction Drucker (1993) believed that the quality of product/service and performance of managers are deciding factors of organizational success. Bass (1990) in a study concluded that 45% to 65% of the total factors that cause success or failure of an organization are decided by leaders. Thus it is important to note that the leadership style of a manager has key relationship with success of an organization. Leadership aims at establishing priorities and marshalling resources to achieve meaningful goals. Managers and supervisors are responsible to determine not only what is important in their organization but also how to get it done by motivating their team members for performing the task. Thus leader must clarify the organizational goals and objectives and to be successful, he must ensure that the team knows both the goals and the strategy. A Study of Relationship between Leader Behaviors and Subordinate Job Expectancies 358 Middle Managers are viewed as critical actors of corporate performance and change. They are responsible for translating the general goals and plans developed by senior management into more specific objectives and activities (Bateman and Snell, 2007). Dopson (1993) proposed that ‘all those below the small group of top strategic managers and above first-level supervision are middle managers’. Middle managers are key players in achieving organizational objectives by motivating employees, removing obstacles, clarifying paths to goal and rewarding them accordingly. Kanter and Stein (1979) rightly quoted that their central position lies in being caught between those below, whose co-operation they need, and those above, who delegate to them the operational authority to implement the stated policy. Traditionally, managers and leaders have been defined independently of each other. It is in recent past that the two roles have been defined in conjunction with each other. Campbell (2004) believes that both terms complement each other and balance the changing needs of organizations. Management ensures order and stability, while leadership produces change and movement. Northouse (2004) argues that both processes .i.e. management and Leadership involve influencing a group of individuals towards goal attainment. 2. Leadership There is a wide range of definitions of Leadership. Stogdill (1974) argued that “there are almost as many definitions of leadership as there are persons who may have attempted to define the concept”. Yukl (2006) believes that “numerous definitions of leadership that have been proposed appear to have little less in common” than involving an influence process. Hemphill and Coons (1957) defined leadership as, “the behavior of an individual when he is directing the activities of a group towards a shared goal”. Tannenbaum et al. (1964) considered leadership influence of human relations. Fiedler (1967) believed that leadership is a process to apply power and influence to make people work together and accomplish common goals. According to Katz and Kahn (1978) leadership is, “the influential increment over and above mechanical compliance with the routine directives of the organization”. Rauch and Behling (1984), Robbins (1993) and Hsieh (1993) regarded leadership as the process of influencing a group to the goal achievement. Jacob and Jaques (1990) defined leadership as “a process of giving purpose or meaningful direction to collective effort and causing willing effort to be expended to achieve purpose”. House et al. (1999) believed leadership as an “ability of an individual to influence, motivate and enable others to contribute towards the effectiveness and success of an organization”. Leadership is not a position, title or privilege rather a process which is observable; understandable that touches on many personal, organizational and social processes. It influences other persons and inspires them to work for the organizational objectives through motivation and not coercion. 2.1 Theories of leadership As the topic of scholarly debate and, the subject of theoretical and empirical research for the past 100 years, leadership has a long tradition. Although many ideas of the past have fallen from popular favor, the evolution of leadership is both reflected in a critical to the Malik 359 understanding of the dominant leadership theories of the present day (Barling et al., 2011). The leadership theories can be classified as; (1) Trait Theories: These theories tried to identify the physical or psychological traits that differentiated leaders from non-leaders and good leaders from bad leaders (House and Aditya, 1997). (2) Behavior Theories: These theories attempted to uncover and verify leadership behaviors that were universally effective. These studies through experiments proved that different leadership styles produce different and complex reactions from same group. Stogdill (1963), Likert (1967), and Kotter (1988) also offered behavioral theories of leadership. (3) Situational Contingency Theories: These theories are build on behavioral theories and believe that effectiveness of leader traits or behavior is dependent upon situation that include organization type, the workplace and the followers (Barling et al., 2011).Fiedler’s (1967) contingency model, House (1971) path-goal theory and Kerr and Jermier (1978) substitute for leadership have received significant theoretical and empirical attention. (4) Modern Approaches to Leadership: Burns (1978) introduced the concept of ‘transforming leadership’ stating “transformational leadership occurs when one or more persons engage with others in such a way that leaders and followers raise one another to higher level of motivation and morality”. Weber (1947) and House (1977) introduced concept of charismatic leadership which became popular in 1980s and 90s. There is no agreement between practitioners and academics as which theory or model is most effective since a single theory cannot explain all situations/circumstances having both strengths and weaknesses. 2.2 Path-Goal Theory Companies employ large number of persons having different ability and motivation. Therefore, it is important for managers to use an appropriate leadership style for subordinate’s readiness to accomplish tasks and achieve organizational goals. Considering the purpose and context of research, House (1974) Path-Goal leadership theory is appropriate model for this study. Hunt (1996) argues that path-goal leadership theory (House, 1974) has existed for almost four decades and it is currently one of the major approaches to leadership that is covered by virtually all basic textbooks on management and organizational behavior Additionally, number of doctoral thesis, over 120 scholarly articles and several in depth reviews have been written exploring the theory’s scientific merits (e.g., Wofford and Liska, 1993; Yukl, 1998). House (1996) in “Path-Goal Theory of Leadership: Lessons, legacy, and reformulated theory” says that path-goal theory is primarily a theory of task and person oriented supervisory behavior which does not concerns the leadership of entire organization, rather only the specified work unit. It specifies leader behaviors that enhance subordinate satisfaction and effectiveness of both work unit and subordinate by addressing the effects of leaders’ behaviors on the motivation and abilities of immediate subordinate and work unit performance. Barling et al. (2011) believe that path-goal theory had two objectives: (a) identify roles and behaviors of effective leaders (b) explore situational contingencies that modify those A Study of Relationship between Leader Behaviors and Subordinate Job Expectancies 360 behaviors. House and Mitchell (1974) said that leader behavior is acceptable and satisfying to the extent that the subordinates see such behavior either an immediate source of satisfaction or instrumental to future satisfaction. Kreitner and Kinicki (2007) believe that leader behavior is predicted to be motivational if: (1) reduces roadblocks that interfere with goal accomplishment (2) provides guidance and support needed by subordinates (3) ties meaningful rewards to goal accomplishment. Thus leader role is to align individual goals with organizational goals and facilitate the achievement of such goals by encouraging them to achieve these goals, clarifying the path towards goal attainment and ensuring that goals are valuable to followers. House and Mitchell (1974) identified four categories of leadership behavior. Directive leader gives subordinates clear and specific instructions to perform their tasks, the timeline for task, and the standards of performance measurement. Supportive leader shows concern for the well being and needs of the subordinates and treat them as equals. Participative leader involves subordinates in decision making by asking for ideas, opinions and takes their suggestions into account. The final leader behavior identified is Achievement-Oriented which involves creating challenging and high standard performance goals for subordinates and seeks for continuous improvement by showing great confidence in subordinates. Eagly and Johnson (1990) believe that leadership behaviors are, by definition, behaviors that fall within a range of typical behaviors but are not always identical or exhibited. Leadership behaviors are not fixed behaviors rather depending on the situation; leaders will vary their behaviors as required by the situation at hand. Path-goal leadership proposes that the effectiveness of leader is influenced by the interaction of leader behaviors (directive, supportive, participative and achievement- oriented) with two types of contingency factors: subordinate characteristics and environment. Contingency factors are the situational variables that cause one leadership style to be more effective than another (Kreitner and Kinicki, 2007). . Subordinate characteristics include: Need for Autonomy: refers subordinate desire to be independent and in self control; Need for Achievement: refers to subordinates’ instinct of striving for and attaining a level of excellence(Feldman, 1999); Locus of Control: is belief of subordinates that they are “master of their own fates” or whatever happens to them in life is a result of “luck, chance, or outside people and events” (Daft, 2008); Perceived Ability: is the extent of the subordinates’ ability to perform tasks and achieve goals. The environmental characteristics include: Task Structure is the extent to which the nature and the requirements of task are specified. It is the degree to which a task , job, work assignment is simple, repetitive and unambiguous (House and Dessler, 1974); Role Ambiguity is experiencing lack of clarity about what is expected of one, how one will be evaluated, and criteria for evaluation (House, 1996). It refers to the degree of uncertainty an employee has about the work role such as duties, authority, allocation of time, relationship with coworkers, directives, policies etc.(Nissa, 2003); Stress refers to body’s biological response to an intense physical, emotional or mental demand/threatening situation placed on it by oneself or others (Ellison, 1998). Malik 361 House (1996) says that path-Goal theory was stimulated by Evan’s (1970) paper, “The effects of Supervisory Behavior on the Path-Goal Relationship” and expectancy theory of motivation. House and Mitchell (1974) define the strategic functions of a leader as: 1. Understanding and stimulating subordinates’ needs for outcomes 2. Enhancing followers’ incentives in order to motivate them for attainment of goals 3. Helping the followers to step forward in order to achieve those incentives 4. Making the followers understand what is expected of them 5. Finally, the leader should reduce those barriers which create frustrations and enhance chances that effective performance results in personal satisfaction. House, (1970) says that based on expectancy theory, leaders should increase the personal rewards, subordinates receive upon achieving goals along with making the path to these goals easier to follow, by clarifying it and reducing roadblocks and pitfalls. White and Bendar (1986) noted that in order to predict specific behavior in a particular situation, individual’s expectancies in that situation should be considered. Expectancy Theory assumes that people are motivated to work when they believe that they can achieve things they want from their jobs. There is a relationship between expectancy, effort, productivity and reward. Werner (2002) states that a person will exert a high effort if he/she believes there is reasonable probability that the effort will lead to the attainment of an organizational goal, and the attainment of the organizational goal will become an instrument through which that person will attain his/her personal goals. According to expectancy theory, motivation depends on a person’s belief that efforts lead to performance (expectancy 1) and performance leads to rewards (expectancy II). House and Dessler (1974) believe that expectancy II refers to the degree to which high quality, quantity and timely performance lead to extrinsic rewards such as increased pay, promotion, recognition or security. According to Coetsee (2003), performance is the realization of goals and meeting of expectations. While Kew et al. (2007) define performance management as a continuous cycle of improving job performance with goal-setting, feedback, coaching, rewards and positive reinforcement. Managers play central role in sustaining employees’ commitment to perform their jobs, boost their morale and ensure job satisfaction. Charlton (2000) believes that extrinsic and intrinsic rewards have reciprocal motivational effects. Similarly, Shah and Shah (2008) summed up that the need for recognition and a sense of belonging affect employees’ perform beyond expectation. Expectancy X Instrumentality X Valence Effort Performance Rewards A Study of Relationship between Leader Behaviors and Subordinate Job Expectancies 362 Indvik (1989) believe that subordinate expectancies (I&II) are the “cornerstones of Path- goal explanation, their absence indicates a dearth of complete tests of path-goal hypotheses”. This study addresses this issue by studying relationship between leader behaviour and subordinate job expectancies. 3. Method The purpose of this study was to investigate the relationship between leader behavior and subordinates job expectancies. This study was dealing with variables whose manifestation had occurred already. Thus Correlational design was used to investigate the effect of leader behavior on subordinates’ job expectancies. 3.1 Participants Participants included 200 employees working in a variety of jobs in cellular industry. Stratified random sampling technique was used and to ensure equal representation from the four selected cellular companies, 15 middle managers and 35 supervisors were randomly selected from the list. 3.2 Procedure Participants were asked their opinion about leader behavior and their own job expectancies through survey questionnaire. The survey included a cover letter informing participants about the purpose of study and requested to complete survey on the basis of their work experience and attitudes. 3.3Measures Leadership Behavior: Four leader behaviors were measured through a set of 20 questions; five each measuring directive and participative leadership behavior; seven measuring supportive; and three measuring achievement-oriented leader behavior. Items were used to measure the perception of participants about their leader behavior. A sample item for supportive leader behavior is, “He is friendly and approachable” and for participative leader behavior, “Before making decisions, he gives serious consideration to what his subordinates have to say”. Participants indicated their responses on a five-point Likert- type scale (1) always to (5) never. Job Expectancies: Tow six-item scales were used to measure job expectancy-I and job expectancy-II. A sample item for job expectancy-I is, “Putting forth as much energy as possible, leads to my producing high quality output” and for job expectancy-II, “The Company gives me recognition for producing high quality output”. Participants indicated the extent to which they believe the outcome using five-point Likert-type scale (5) strongly agree to (1) strongly disagree. 4. Results The results of study revealed that gender, age, educational qualification and experience did not affect job expectancies of participants except the position/hierarchy i.e. supervisors and middle managers had different perceptions of job expectancies (I&II).While, Leader behavior is significantly related with subordinates’ job expectancies(I&II). Malik 363 H01. Subordinates’ attributes (age, gender, qualification, rank, experience and length of service under the current supervisor) strongly affect their job expectancies (I&II). Predictors Β t –value p R F-Ratio Sig R2 Age .957 .063 .512 .019 1.057 0.395 0.54 Gender 1.509 .116 .193 .138 Qualification .205 .049 .576 .043 Rank 2.098 .161 .076 .144 Experience -.295 -.045 .667 -.002 Service -1.137 -.123 .172 -.097 The R-square of .54 implies that the attributes of subordinates accounted for 54 percent of the variation in subordinates’ job expectancy I and value of F (1.057) was not significant. Similarly, beta and t-values of age , gender, qualification, rank, experience and length of service were all insignificant and had moderate correlation except for experience and service under current supervisor ( having -ve correlation) with the subordinates job expectancy I. Predictors Β t – value p R F- Ratio Sig R2 Age -1.034 -.696 .488 -.063 1.453 0.190 0.73 Gender .015 .013 .990 .030 Qualification -.309 -.826 .410 -.064 Rank 3.367 2.865 .005 .246 Salary .246 .634 .527 .035 Experience -.252 -.361 .719 .000 Service -.293 -.348 .728 .004 The analysis of data resulted in R-square of .73 while value of F (1.453), p (.190) was not significant that means the attributes of subordinates cannot be used as predictors of job expectancy II. A Study of Relationship between Leader Behaviors and Subordinate Job Expectancies 364 H02. Leadership behavior significantly affects subordinates’ job expectancies (I and II). Count Correlation p value R-Square % VOC Directive 156 .062 .439 .003 0.38 Supportive 158 .159 .046 .025 2.50 Participative 158 .036 .652 .001 0.12 Achievement Oriented 158 .129 .105 .016 1.66 The correlation coefficient of leader behavior and job expectancy I is (.062); (.159); (.036) and; (.129) for directive, supportive, participative and achievement-oriented respectively. The three correlation coefficients (directive, participative and achievement oriented leader behavior) with expectancy I are not significant. Somehow, there is a weaker relationship between the supportive leadership behavior and job expectancy I, as only 2.5% variance in job expectancy I is due to supportive leadership behavior. Count Correlation p value R-Square % VOC Directive 156 .091 .257 .008 0.82 Supportive 158 .078 .330 .006 0.60 Participative 158 .034 .672 .001 0.11 Achievement Oriented 158 .110 .169 .012 1.12 The correlation coefficient of expectancy II & leader behavior is (.091); (.078); (.110) for directive, supportive and; for achievement –oriented leader respectively which are not significant while, correlation coefficient for participative leadership behavior and expectancy II is (.034) which is significant and only 0.11% variance in job expectancy II is due to participative leadership behavior. Malik 365 The linear regression analysis for linear combination of four leader behaviors and job expectancies is as follows t value P Β F-Ratio R R2 7.919 .000 18.122 2.618 .128 .017 Based on the values of job expectancy-I (.128), t (158) =7.919, and p =.000, It was concluded that leadership behavior affects subordinates perception of subordinates that effort leads to performance (job expectancy-I). t value P Β F-Ratio R R2 7.488 .000 17.620 1.419 .095 .009 While, the correlation between leadership behavior (directive, supportive, participative and achievement-oriented) and job expectancy II was .095. The R-square of .009 implies that .9 percent of the variation in job expectancy II. H03. Situational factors (task structure, role ambiguity, stress, need for autonomy, locus of control, need for achievement and perception about abilities) affect subordinates’ job expectancies (I & II). Predictors Β t – value p r F- Ratio Sig R2 Locus of Control -.022 -.133 .894 .238 7.825 .000 .317 Ability .419 2.380 .019 .434 Task Structure -.169 -1.554 .123 -.037 Role Ambiguity .247 3.104 .002 .387 Stress .014 .138 .890 -.002 Achievement Need -.035 -.244 .808 .222 Autonomy Need .546 2.669 .009 .437 The value of square (.317) implies that the situational factors accounted for 31 percent of the variation in job expectancy- I. The value of F (7.825) was greater than critical value. Further, based on t values it was concluded that situational factor: role ambiguity; autonomy need and; ability could be used as predictor of job expectancy I. A Study of Relationship between Leader Behaviors and Subordinate Job Expectancies 366 Predictors Β t – value p r F- Ratio Sig R2 Locus of Control .210 1.159 .249 .276 5.463 .000 .245 Ability -.161 -.851 .396 .160 Task Structure -.049 -.421 .674 .088 Role Ambiguity .295 3.454 .001 .356 Stress -.107 -.944 .347 -.074 Achievement Need -.231 -1.505 .135 .082 Autonomy Need .664 3.016 .003 .391 The R-square of .245 implies that the situational factors accounted for 24 percent of the variation in expectancy II. Based on value of F (5.463) which was greater than the critical value it was further concluded that individual variables: role ambiguity t (126) = 3.454, p = .001 and; autonomy need t (126) = 3.016, p = .003 can be used as predictor of subordinates’ job expectancy II. 5. Findings The Null Hypothesis of the study that there is no significant relationship between leader behavior and subordinates’ job expectancies was accepted. While the Null Hypothesis that there is no significant relationship between linear combination of four leader behaviors(directive, supportive, participative and achievement-oriented) and subordinates’ job expectancies (I&II) of subordinates was rejected. Thus it may be deduced that leader behavior affects subordinates’ job expectancies and can be used as predictor of subordinates’ job expectancies. Further, Null Hypotheses that there is no significant relationship between subordinates attributes (age, gender, qualification, rank, experience and length of service under the current supervisor.) and job expectancies (I&II) was accepted except rank/position and expectancy-II, which was rejected. The Null Hypothesis that there is no significant relationship between situational factors (task structure, role ambiguity, stress, need for autonomy, locus of control, need for Malik 367 achievement, and perception about abilities) and subordinates’ job expectancies (I&II) was rejected. The regression analysis for situational factors (locus of control, ability, task structure, role ambiguity, stress, achievement need and autonomy need) and job expectancy I had strong R –square (.317) which implies that the situational factors accounted for 31 percent of the variation in expectancy I while F value (126) 7.825 was greater than critical value(.05). Similarly, the Null Hypothesis that there is no significant difference in the perception of Supervisor/lower manager and middle manager regarding job expectancy (I&II) was rejected. 5.1 Discussion The major objective this study was to test path-goal theory with reference to expectancy theory. The reward management system of the cellular companies gives enough powers to its managers to reward the high performing individuals but this study revealed a weaker relationship between leader behavior and subordinates’ job expectancy (I) except for supportive leader behavior. This is in non conformity with path-goal assumption that superior behavior may increase effort when it makes satisfaction of a subordinates’ needs contingent upon effective performance and/or when it complements work environment through coaching, guidance, support and rewards necessary for effective performance. According to House(1971, p.234), “the motivational function of the leader consists of increasing payoffs to subordinates for work-goal attainment and making the path to these pay offs easier to travel by clarifying it, reducing roadblocks and pit falls, and increasing the opportunities of personal satisfaction en route”. While linear combination of leader behavior affected the subordinates job expectancies. This result points out a potential flaw in the management practices in the judicious use of reward system. When an individual is convinced that organizational justice does not prevail and there is no difference between high performer and poor performer rather both are being treated equally, then individuals stop putting extra efforts and giving high productivity. Thus it has important implications for the HR managers. This is further supported by expectancy theory which describes work motivation in terms of a rational choice process in which a person decides how much effort to devote to do the job at a given point of time. In choosing between maximal/minimal efforts, a person considers the likelihood that task completion will result in desirable outcomes (Vroom, 1964). Subordinates’ attributes/characteristics (age, gender, qualification, rank, experience and length of service under the current supervisor) don’t affect job expectancies (I&II) except rank/position and expectancy-II, which means that every individual is very well aware of the existing reward system which educates organizational members as if what is expected from them and as result what they can expect from the system. The situational factors (task structure, role ambiguity, stress, need for autonomy, locus of control, need for achievement and perception about abilities) affect subordinates’ job expectancies (I&II).Path goal theory predicted that directive leader behavior will be more effective for the subordinates with high need for achievement because directive leader through clarifying path guides subordinates. Similarly, participative leader behavior is also effective as he consults with subordinates in setting, clarifying and achieving goals. The results of this study reveal that there is an inverse relationship between subordinates’ A Study of Relationship between Leader Behaviors and Subordinate Job Expectancies 368 job expectancy (I&II). According to Yukl (2006), for subordinates with high need for autonomy, participative leader behavior will increase the intrinsic valence of work. The … The Leadership Quarterly 25 (2014) 544–562 Contents lists available at ScienceDirect The Leadership Quarterly journal homepage: www.elsevier.com/locate/leaqua Same difference? Exploring the differential mechanisms linking servant leadership and transformational leadership to follower outcomes Dirk van Dierendonck ⁎, Daan Stam, Pieter Boersma, Ninotchka de Windt, Jorrit Alkema Rotterdam School of Management, Erasmus University, The Netherlands a r t i c l e i n f o ⁎ Corresponding author at: Rotterdam School of Man fax: +31 4089015. E-mail address: [email protected] (D. van 1048-9843/$ – see front matter © 2013 Elsevier Inc. A http://dx.doi.org/10.1016/j.leaqua.2013.11.014 a b s t r a c t Article history: Received 26 July 2012 Received in revised form 26 September 2013 Accepted 18 November 2013 Available online 9 December 2013 Handling Editor: Shelly Dionne work engagement; however, the manner in which they exerted their influence differed. SL This paper aimed to provide insights into the different mediating mechanisms through which servant leadership (SL) and transformational leadership (TFL) affect followers. We also investigated environmental uncertainty as a moderator of the effects of servant leadership and transformational leadership. Based on the results of two experimental studies and one field study, we concluded that both SL and TFL were related to organizational commitment and worked primarily through follower need satisfaction, whereas TFL worked mainly through perceived leadership effectiveness. The moderating influence of uncertainty was inconsistent across the studies. © 2013 Elsevier Inc. All rights reserved. Keywords: Servant leadership Transformational leadership Engagement Commitment 1. Introduction Leadership as a topic in management has generated an abundance of research over the past several decades. Although it would be going too far to suggest that leadership scholars agree on which behaviors and styles are optimal for leadership, it is clear that one style, more than any other, has been found to be effective: transformational leadership (TFL) (see the meta-analysis by Judge & Piccolo, 2004). However, as business environments change, leadership may face new challenges. One particularly important trend in this respect is the growing dependency on people in a knowledge-based economy, which makes attention to the needs of employees essential for long-term success (O'Leary, Lindholm, Whitford, & Freeman, 2002). Consequently, scholars have recently investigated a type of leadership that is particularly oriented to the needs of employees, known as servant leadership (SL), and although research on SL is in a relatively early stage, empirical findings regarding SL are promising (Van Dierendonck, 2011). However, several scholars have emphasized the considerable overlap between SL and TFL. An important aspect for research on SL is, therefore, investigating whether SL is actually different from TFL and, if so, how. Several theoretical papers have argued that SL and TFL have different foci and may be suitable to different environments; TFL focuses on organizational effectiveness, whereas SL focuses on follower needs (Bass, 1985; Graham, 1991, 1995; Mayer, Bardes, & Piccolo, 2008; Van Dierendonck, 2011). Graham (1991) argued that where TFL emphasizes the leader's skills, hierarchical power relationships between leader and follower, visions for the organization, and especially performance, effort, and achieving the goals set out by the leader, SL emphasizes the humility and spirituality of leaders, mutual power, visions of a way of life for the leader and followers, emulation of the leader's service orientation, and the autonomy and moral development of followers. Graham (1995) added that while SL accomplishes OCB among followers by causing followers to reason in terms of universal agement, Erasmus University, Burg Oudlaan 50, 3062 PS Rotterdam, The Netherlands. Tel.: +31 4089569; Dierendonck). ll rights reserved. 545D. van Dierendonck et al. / The Leadership Quarterly 25 (2014) 544–562 principles and justice, TFL does so by applying to utilitarian calculus and cost–benefit analysis for stakeholders. Smith, Montagno, and Kuzmenko (2004) also emphasized that while TFL focuses on change and organizational innovation and is especially effective in times of uncertainty, SL seems more oriented on preserving the status quo and focusing on individual people and is especially effective in time of stability (cf. Choi & Mai-Dalton, 1998). This suggests that SL and TFL affect outcomes through different processes (TFL through processes related to organizational effectiveness and SL through processes related to follower need satisfaction) and are effective under different circumstances (TFL would be especially effective under uncertainty, while SL would be especially effective under stability). Unfortunately, however, the few empirical studies that have investigated these fundamental differences have gone no further than establishing the divergent validity of SL and TFL and demonstrating that SL explains unique variance in outcomes (such as follower commitment) beyond the effects of TFL (see, for instance, Barbuto & Wheeler, 2006; Liden, Wayne, Zhao, & Henderson, 2008; Peterson, Galvin, & Lange, 2012). Although these studies answer the question of whether SL and TFL are different, to test the fundamental differences between SL and TFL (i.e., how TFL and SL are different), there is a need for empirical research that goes beyond the correlation between SL and TFL and directly investigates the different underlying mechanisms through which these forms of leadership affect outcomes and the different boundary conditions for these effects. This is what the current research aims to accomplish. In a series of three studies using a variety of research methods (experiments and survey research), we investigated the relationships between SL and TFL and between commitment and engagement. Importantly, we study whether the effects of TFL are mediated by followers' perception of leadership effectiveness and whether the effects of SL are mediated by followers' need satisfaction. Moreover, we test the moderating effect of times of uncertainty, specifically, whether the effects of TFL are stronger and those of SL are weaker in times of greater uncertainty. By going beyond investigations of whether SL and TFL are different constructs and investigating directly how SL and TFL are different, the current research aims to contribute to the leadership literature in various ways. First, prior research has mainly focused on the discriminant validity of measures of SL and TFL. However, the fact that measures of two concepts can be discriminated does not imply that the theoretical concepts are different. The current research provides a first step to differentiating between SL and TFL based on theoretical models that specify mediating and moderating pathways of effects of SL and TFL and as such provides much-needed support for the notion that SL and TFL are indeed different theoretical concepts that work through different processes. Second, comparing the mechanisms underlying the effects of SL and TFL provides a much better understanding of why these leadership styles are effective. Consequently, this provides the field with insights into when and where the different leadership styles can be optimally effective (i.e., ideas for moderators to investigate). For instance, the finding that need satisfaction underlies the effectiveness of SL implies that in environments in which need satisfaction is a constraint and cannot possibly be expected to change, SL may not be very effective. Thus, an understanding of the differential mechanisms underlying SL and TFL can form the basis for theorizing about boundary conditions for these styles. In the current manuscript, the moderation of uncertainty is an example of such novel theorizing. Third, our research also has practical value. By showing that SL and TFL work through different processes, we provide information to managers and companies that may affect their decision to promote one style over the other. Organizations that emphasize need satisfaction may choose SL as their preferred leadership style for managers, while organizations that are more oriented to perceptions of effectiveness may prefer TFL. Our research provides the basis for more evidence-based decisions concerning management styles. In the following, we first detail what TFL and SL refer to, and we subsequently develop a comprehensive conceptual model of the process through which these leadership styles affect the commitment and engagement of followers. Finally, we discuss the studies conducted. 2. Transformational leadership and servant leadership TFL refers to a multidimensional leadership style that encourages followers to perform beyond expectations and emphasizes collective values and needs rather than followers' individual values and needs (Bass, 2005; Yukl, 1999). The different definitions of TFL have a common primary focus on organizational goals: transformational leaders inspire their followers to perform better for the sake of the organization. Rewards and praise are used to encourage a stronger focus on achieving high outcomes (Rafferty & Griffin, 2004). TFL theory discusses various elements (or dimensions) of leader behaviors. For instance, Bass (1985) includes inspirational motivation (communicating a stimulating vision), idealized influence (serving as a motivating role model), intellectual stimulation (stimulating followers to think outside of the box), and individualized consideration (an emphasis on followers' development). Rafferty and Griffin (2004) add personal recognition (recognizing the performance of followers) to these elements. Although often contrasted with transactional leadership (a leadership style that emphasizes the exchange relationship between leaders and followers and focuses on explaining and setting goals and providing rewards/punishment), TFL is not the opposite of transactional leadership but instead a leadership style that surpasses explanations, goal setting, and providing rewards for follower performance (Bass, 1985). TFL is generally viewed as an effective leadership style, and studies show that TFL has many positive effects. For instance, TFL positively predicts work motivation (Shamir, House, & Arthur, 1993), employee satisfaction (Podsakoff, MacKenzie, Moorman, & Fetter, 1990), the number of accidents in warehouses (De Koster, Stam, & Balk, 2011), and innovative performance (Nederveen Pieterse, Van Knippenberg, Schippers, & Stam, 2010). For overviews of the effects of TFL, see Lowe and Kroeck (1996) and Bass and Riggio (2006). 546 D. van Dierendonck et al. / The Leadership Quarterly 25 (2014) 544–562 Notably, after decades in which TFL was the most widely studied leadership style, SL has been receiving increased attention in the leadership field (Van Dierendonck, 2011). The literature on SL advocates that servant leaders must primarily meet the needs of others (Greenleaf, 1977). Servant leaders focus on developing employees to their fullest potential in areas of task effectiveness, community stewardship, self-motivation, and future leadership capabilities, and they provide vision and gain credibility and trust from followers (Farling, Stone, & Winston, 1999). To generate the best performance in their followers, servant leaders rely on one-on-one communication to understand the abilities, needs, desires, goals, and potentials of their followers. With knowledge of each follower's unique characteristics and interests, leaders then assist followers to achieve their potential (Liden et al., 2008). Barbuto and Wheeler (2006) described SL as including an altruistic calling, which is the motivation of leaders to place others' needs and interests ahead of their own, and organizational stewardship, which orients others toward benefiting and serving the community. Recent research on SL has shown promising results for its basic premises. For instance, Van Dierendonck and Nuijten (2011) found a relationship between SL and job satisfaction and work engagement (using various different samples, including high school staff, civil servants, and gas station employees). Other research has found relationships with trust and team performance (Schaubroeck, Lam, & Chunyan Peng, 2011), organizational citizenship behavior (Walumbwa, Hartnell, & Oke, 2010), team potency (Hu & Liden, 2011), and firm performance (Peterson et al., 2012). For an overview of the SL research, see Van Dierendonck (2011). The descriptions of TFL and SL emphasize that there is considerable overlap between the two leadership styles. Both transformational and servant leaders are focused on their followers, extend leadership beyond simply setting and explaining task goals, provide visions for the future, and are generally positively correlated with various important outcome measures. However, we assert that there may be major differences in the way that they influence their followers and in the extent that their effectiveness is influenced by the environment. We will elaborate on these differences in the next section. 3. The differential influences of servant leadership and transformational leadership As formulated by Stone, Russell, and Patterson (2004. P. 1) ‘The extent to which the leader is able to shift the primary focus of leadership from the organization to the follower is the distinguishing factor in classifying leaders as either transformational or servant leader’. Barbuto and Wheeler (2006) suggested that another important difference between SL and TFL is that SL focuses on a desire to serve and preparing others to serve as well, whereas TFL emphasizes a desire to lead and inspiring followers to perform well. Empirical research into these differences between SL and TFL has mostly been performed in the process of scale development. For instance, in their scale development efforts, Barbuto and Wheeler (2006), Liden et al. (2008), and Van Dierendonck and Nuijten (2011) were able to show that for their respective instruments, SL could be meaningfully and statistically distinguished from TFL. Regretfully, none of these studies tests the fundamental theoretical differences between SL and TFL as proposed by early theorizing (see also Graham, 1991, 1995; Smith et al., 2004). In the following, we focus first on the different mechanisms underlying TFL and SL (i.e., attributions of leadership effectiveness and psychological needs) and how these affect outcomes (engagement and commitment of followers). We chose to focus on the above two mechanisms because they are suggested as the primary mechanisms that underlie the effects of TFL and charisma (in the case of perceptions of leadership effectiveness; Sy, Choi, & Johnson, 2014) and SL (in the case of psychological needs, Mayer, 2010) and as such should be the most distinguishing mechanisms. Next, we address how the environment (i.e., uncertainty) can differentially moderate the effects of TFL and SL. We chose uncertainty as a moderator because research suggests opposite moderation effects of uncertainty on the effects of TFL (positive moderation) and SL (negative moderation), thereby providing us with a moderator that optimally distinguishes between TFL and SL. 3.1. Attributions of leadership effectiveness Perceptions of followers concerning their leaders' effectiveness play an important role in leadership (cf. Meindl, Ehrlich, & Dukerich, 1985). Attributions of leadership effectiveness are important because they provide followers with a sense of trust in their leader; individuals are more likely to follow leaders who they believe to be competent and effective. In this sense, the perception of leadership effectiveness represents the overall evaluation of leader more than any other possible mediating mechanism. We argue that TFL more strongly affects such attributions than SL. We base this proposition on several streams of research. First, whereas TFL places the leader at the center of the leadership process, SL emphasizes the follower. TFL inherently has a charismatic component (Bass, 1985). By communicating a stimulating vision and acting as role models, transformational leaders often lead followers to attribute charisma and leadership effectiveness to leaders. It is through processes of vicarious learning and personal identification (Yukl, 1999) that followers of transformational leaders learn the norms and values of the organization. The high correlations between TFL and leader effectiveness in the meta-analysis of Judge and Piccolo (2004) are consistent with this reasoning. Moreover, Sy and colleagues (2014) recently showed a reciprocal relationship between charisma and leader effectiveness perceptions in a longitudinal study. SL, on the other hand, is less leader-focused than TFL (cf. Graham, 1991, 1995). TFL puts the leader at the center of the group, while servant leaders will attribute successes to followers instead of themselves. When considering the various scales of SL, it is telling that many incorporate dimensions such as altruistic calling (Barbuto & Wheeler, 2006), putting subordinates first (Liden et 547D. van Dierendonck et al. / The Leadership Quarterly 25 (2014) 544–562 al., 2008), and standing back and humility (Van Dierendonck & Nuijten, 2011). Although such behaviors may be crucial to developing followers' feelings of competence and impact, they may come at the expense of being observed as extra-ordinary and a main source of performance. Thus, we suggest that TFL is highly visible and uses influence processes that rely on the leader taking the stage, whereas SL is less visible and uses influence processes that allow SL leaders to be more in the background. Second, research on the romance of leadership (Meindl et al., 1985; Meindl & Ehrlich, 1987) argues that due to their notable position, leaders are generally attributed with more success (and failure) than would be predicted based on their behaviors. Interestingly, people even consider leader-attributed performance to be more positive than performance attributed to non-leader factors (Meindl & Ehrlich, 1987). In other words, leaders are accredited with more performance influence than is reasonable, and such performance is also observed as more positive than objectively appropriate. TFL uses this process to its advantage. By putting the leader center stage, the romantic image of leadership is optimally used to create an aura of effectiveness and charisma that is TFL's main source of influence. In other words, in addition to being very visible, group performance is even more highly attributed to TFL leaders due to this visibility. On the other hand, SL actively tries to remedy the notion of romance of leadership by showing humility, standing back, and generally emphasizing that it is the followers, rather than the servant leader, that are the cause of performance. To the extent then that they succeed, followers may therefore be less inclined to attribute the group's successes to the leader. Thus, SL is not only less visible than TFL, but this lack of visibility (combined with the leader's attribution of successes to followers) may also translate into SL leaders being less accredited for the group's successes. This leads to the following hypothesis: Hypothesis 1. Transformational leadership is more strongly related to follower perceptions of leadership effectiveness than servant leadership. 3.2. Psychological needs The fulfillment of basic psychological needs is a key determinant of health and well-being (Ryan & Deci, 2000). Many such psychological needs exist, but a dominant theory in the field, self-determination theory, suggests that there are three crucial elements of psychological needs for people (Ryan & Deci, 2000). Competence refers to effectively acting on and influencing one's environment. Autonomy is the experience of one's own will and initiative in one's own behavior. Relatedness refers to feelings of connection and belonging. The need for competence can be satisfied by offering optimal challenges and providing relevant feedback (Ryan & Brown, 2003). Autonomy is achieved through the informal control of employees by the leader and the perception of the freedom of choice on behalf of the individual. The need for relatedness is satisfied when the individual experiences warmth, acceptance, and care. The fulfillment of these needs can enhance an individual's intrinsic motivation and result in a sense of self-determination. Concern for the needs of followers is more strongly emphasized in SL theory than in any other leadership theory (Mayer, 2010). Servant leaders invest time and energy to understand the needs of all individual followers and subsequently work to satisfy these needs. It is the core element of the famous quote by Greenleaf (1977) on what he considers the true test of servant leadership: ‘The difference manifests itself in the care taken by the servant-first to make sure that other people's highest priority needs are being served. The best test, and difficult to administer, is: Do those served grow as persons?’ There is also empirical support for the relationship between SL and need satisfaction (Mayer et al., 2008). In TFL theory, personal attention for followers is aimed at enhancing innovation and creativity (Smith et al., 2004) to ultimately reach another – organizational – goal. In SL, personal attention is given for its own sake, to help followers grow as persons, and organizational goals are secondary (Greenleaf, 1977). Graham (1995) suggests that although TFL, similar to SL, encourages constructive participation, the level of moral development is primarily utilitarian and focused on the costs and benefits of the stakeholders. In contrast, SL comes from recognition of universal principles and a focus on justice and the greater whole. Indeed, while TFL incorporates elements related to a focus on individuals overall, the style strongly emphasizes the organization. This alignment is not problematic when the individuals' and organization's needs are in line, and organizational needs are deemed more important when the individuals' needs otherwise. Servant leaders would put the needs of followers first, even if they would conflict with their personal or the organization's interests. This leads to the following hypothesis: Hypothesis 2. Servant leadership is more strongly related to the satisfaction of the psychological needs of followers than transformational leadership. 3.3. Organizational commitment and work engagement Commitment can be broadly defined as “a force [that] binds an individual to a course of action that is of relevance to a particular target” (Meyer & Herscovitch, 2001, p. 301). Although different forms of commitment can be observed in studies of organizational commitment (i.e., affective, continuance, and normative commitment; see Allen & Meyer, 1990), affective commitment tends to be the most relevant form for predicting positive organizational behaviors (Mathieu & Zajac, 1990). This result has been confirmed by several meta-analyses with a broad range of outcomes, including job performance, turnover, absenteeism, and organizational citizenship behaviors (OCBs) (e.g., Mathieu & Zajac, 1990; Meyer, Stanley, Herscovitch, & Topolnytsky, 2002; Riketta & Van Dick, 2005). Affective organizational commitment can be observed as “an emotional attachment 548 D. van Dierendonck et al. / The Leadership Quarterly 25 (2014) 544–562 to, identification with, and involvement in the organization” (Meyer et al., 2002: 21). Similar to affective commitment, work engagement also refers to an attachment to, identification with, and involvement in an object or activity, but in this case, the object of this attachment is the work itself rather than the organization. Work engagement has been defined as consisting of three dimensions: dedication to work activities, absorption in work activities, and vigor in the pursuit of work activities (Schaufeli, Bakker, & Salanova, 2006). Prior research has shown that TFL enhances organizational commitment (Avolio, Zhu, Koh, & Bhatia, 2004; Bono & Judge, 2003; Dvir, Eden, Avolio, & Shamir, 2002) and work engagement (Zhu, Avolio, & Walumbwa, 2009). Studies on SL have similarly suggested that SL fosters work engagement and affective commitment. For instance, research has shown that SL positively predicts work engagement (Van Dierendonck & Nuijten, 2011), while Liden et al. (2008) demonstrate that SL strongly predicts organizational commitment. This leads to the following hypothesis: Hypothesis 3a. TFL and SL both positively influence organizational commitment and work engagement. We maintain that two pathways through which SL and TFL affect commitment and engagement of followers are attributions of leader effectiveness and satisfaction of psychological needs. Perceptions of leader effectiveness provide followers with the feeling that their leaders and organizations are competent and worthy, and such positive evaluations may foster engagement and commitment on the part of followers. Indeed, research has shown that perception of effectiveness (and charisma) may positively affect such outcomes as commitment and engagement (cf. DeGroot, Kiker, & Cross, 2000). Furthermore, research on need satisfaction indicates that need satisfaction is positively related to organizational commitment (Brown, 1969; Hall, Schneider, & Nygren, 1970) and work engagement (Deci, Ryan, Gagne, Usunov, & Kornazheva, 2001; Van Den Broek, Vansteenkiste, De Witte, & Lens, 2008) because followers reciprocate the need satisfaction that they experience. Combining these findings with the idea that SL and TFL influence perceptions of leadership effectiveness and satisfaction of the psychological needs of followers suggests that the latter may indeed be mediators of the relationship between SL and TFL as well as follower commitment and engagement. Considering that TFL is a strong predictor (and stronger than SL) of attributions of leadership effectiveness and that SL is a strong predictor (and stronger than TFL) of the need satisfaction of followers, we suggest that the mediation of attributions of leader effectiveness is especially strong for TFL, while the mediation of need satisfaction is especially strong for SL. Hypothesis 3b. Attributions of leader effectiveness mediate the influence of TFL and SL on organizational commitment and work engagement, but the mediation of the influence of TFL is much stronger than the mediation of the influence of SL. Hypothesis 3c. Followers' psychological need satisfaction mediates the influence of SL and TFL on organizational commitment and work engagement, but the mediation of the influence of SL is much stronger than the mediation of the influence of TFL. 3.4. Moderating effect of uncertainty So far, we have argued that SL and TFL work through different processes (attributions of leadership effectiveness versus psychological need satisfaction), but that both affect important outcomes positively. These differential pathways are crucial to empirically studying the fundamental, theoretical differences between SL and TFL (how they work differently). However, understanding these different mechanisms is not only important for theory concerning SL and TFL but may also add important insight into the circumstances under which SL and TFL may be more (or less) effective. Here, we argue that, based on the differential mediating pathways of SL and TFL, an important moderator of the effects of SL and TFL is environmental uncertainty. Importantly, we argue that uncertainty strengthens the effects of TFL but weakens the effects of SL. Let us explain these ideas. When times are uncertain, for instance, during economic crises, people experience high risk and turbulence (Antonakis, Avolio, & Sivasubramaniam, 2003), which may generate feelings of distress and uncertainty (Stubbart, 1987) and undermine people's feelings of safety (Lane & Klenke, 2004). A consequence of feelings of uncertainty and its associated negative affect is that people aim to reduce this uncertainty. There are several ways in which they may do so. First, uncertainty identity theory (Hogg, 2007) argues that people use associations with social groups to reduce uncertainty. Groups provide safety, resources, and support; therefore, in periods of high uncertainty, people seek out groups to associate with or identify more strongly with groups they are attached to. Consequently, as followers tend to focus on groups and organizations more than themselves for comfort and safety in times of uncertainty, we argue that collective, organizational needs also become more pronounced for followers, while personal needs become less salient (cf. Hogg, 2007). Psychological need satisfaction therefore becomes more oriented toward the satisfaction of organizational needs (as opposed to personal needs). Second, in times of uncertainty, the conventional routes of behavior become obsolete and ineffective; some even define crises and uncertainty as events that are ‘…characterized by ambiguity of cause, effect, and means of resolution, as well as by a belief that decisions must be made swiftly’ (Pearson & Clair, 1998, p. 60, see also, Dutton, 1986; Madera & Smith, 2009; Pearson & Mitroff, 1993). Leader effectiveness … Daily Job Demands and Employee Work Engagement: The Role of Daily Transformational Leadership Behavior Kimberley Breevaart and Arnold B. Bakker Erasmus University Rotterdam Using job demands–resources (JD-R) theory, the present study integrates the challenge stressor- hindrance stressor framework and leadership theory to investigate the relationship between daily transformational leadership behavior and employee work engagement. We hypothesized that daily transformational leadership behavior (a) sustains employee work engagement on days characterized by high challenge job demands, and (b) protects work engagement on days characterized by high hindrance job demands. Teachers filled out a short online questionnaire at the end of each workday during a 2-week period (N � 271 � 5.68 days � 1539). Results of latent moderated structural equation modeling showed that teachers’ daily challenge demands (workload and cognitive demands) had a positive relationship with work engagement on the days transformational leadership was high (vs. low). In addition, teachers’ daily hindrance demands (role-conflict, but not family to work conflict) had a negative relationship with work engagement on the days transformational leadership was low (vs. high). These findings show that the function of transformational leadership behavior changes from day to day, and depends on the type of job demand. We discuss the practical and theoretical implications of these findings. Keywords: challenge stressor-hindrance stressor framework, diary study, JD-R theory, transformational leadership, work engagement With their charisma and attentiveness to subordinates’ needs and opportunities for personal development, transformational lead- ers can have a significant impact on the functioning of their subordinates (Bass, 1985). Specifically, subordinates of transfor- mational leaders are generally motivated and satisfied with their work, and show above average performance (for meta-analyses, see, e.g., Judge & Piccolo, 2004; Wang, Courtright, & Colbert, 2011). It is therefore not surprising that transformational leader- ship theory (Bass, 1985, 1999) has attracted enormous research attention. However, it seems unlikely that leaders are able to inspire and challenge their subordinates all the time to the same extent. That is, the use of transformational leadership behavior requires effort, time, and some level of self-control (Furtner, Baldegger, & Rauthmann, 2013), which are known to be finite resources (Baumeister, Bratslavsky, Muraven, & Tice, 1998). Therefore, it is not just important to know whether transforma- tional leadership behavior is used, but also when this behavior is best used. Accordingly, the central aim of the present quantitative diary study is to further our understanding of transformational leadership and to guide managers in motivating their subordinates by examining when transformational leadership behavior is needed the most. On the basis of job demands–resources theory (JD-R theory; Bakker & Demerouti, 2014; Demerouti, Bakker, Nachreiner, & Schaufeli, 2001), we propose that transformational leadership be- havior such as being supportive of subordinates’ needs and inspir- ing subordinates with an optimistic vision of the future (Bass, 1985), acts (refers to “behavior”) as a valuable job resource that is particularly important on days when job demands are high. The challenge stressor-hindrance stressor framework (Cavanaugh, Bo- swell, Roehling, & Boudreau, 2000) states that all job demands consume energy, but that hindrance demands thwart personal growth and goal achievement, whereas challenge demands have the potential to contribute to learning and achievement. Indeed, research has shown that both types of demands are positively related to strain and that challenge demands contribute to em- ployee motivation and performance, whereas hindrance demands are detrimental to employee motivation and performance (LePine, Podsakoff, & LePine, 2005). Using this framework, we propose that transformational leadership behavior fosters employee work engagement on the days that subordinates are confronted with high challenge demands (i.e., workload and cognitive demands), and that transformational leadership behavior sustains work engage- ment on the days that subordinates are confronted with high hindrance demands (i.e., role-conflict and strain-based family to work conflict). Our study makes two significant contributions to the literature. First, we try to advance transformational leadership theory. Most research on transformational leadership focuses on between- person differences, assuming that leaders are either transforma- tional or not (i.e., trait transformational leadership), ignoring pos- sible within-person differences in transformational leadership. Adopting a dynamic approach to leadership, we theorize that there are meaningful fluctuations in transformational leadership behav- ior within the same leader and examine when this behavior is needed the most, that is, how leaders can optimally allocate their This article was published Online First March 30, 2017. Kimberley Breevaart and Arnold B. Bakker, Department of Work and Organizational Psychology, Erasmus University Rotterdam. Correspondence concerning this article should be addressed to Kimber- ley Breevaart, Department of Work and Organisational Psychology, Eras- mus University Rotterdam, Woudestein Campus, room T13-20, P.O. Box 1738, 3000 DR Rotterdam, the Netherlands. E-mail: [email protected] T hi s do cu m en t is co py ri gh te d by th e A m er ic an P sy ch ol og ic al A ss oc ia ti on or on e of it s al li ed pu bl is he rs . T hi s ar ti cl e is in te nd ed so le ly fo r th e pe rs on al us e of th e in di vi du al us er an d is no t to be di ss em in at ed br oa dl y. Journal of Occupational Health Psychology © 2017 American Psychological Association 2018, Vol. 23, No. 3, 338 –349 1076-8998/18/$12.00 http://dx.doi.org/10.1037/ocp0000082 338 resources. Second, we use the challenge-hindrance stressor frame- work to fine-tune JD-R theory. We argue that the resources pro- vided by the leader (i.e., transformational leadership behavior) boost the relationship between challenging job demands (i.e., workload, cognitive demands) and employees’ work engagement, and buffer the relationship between hindrance demands (i.e., fam- ily to work conflict, role-conflict) and employees’ work engage- ment. Static Versus Dynamic Leadership Leadership researchers have always been intrigued by the search for the most effective way or person to lead. It is therefore not surprising that transformational leadership theory has received a great deal of scientific attention. Research interest in transforma- tional leadership gained momentum when Bass (1985), building on the work of Burns (1978), introduced his ideas on transformational leadership. Accordingly, transformational leaders are role models (i.e., idealized influence) who inspire and motivate their followers’ (i.e., inspirational motivation), are genuinely concerned with their followers’ needs (i.e., individualized consideration), and encour- age their followers to be creative (i.e., intellectual stimulation). No fewer than five meta-analytic studies support the effectiveness of transformational leadership (e.g., DeGroot, Kiker, & Cross, 2000; Judge & Piccolo, 2004; Lowe, Kroeck, & Sivasubramaniam, 1996; Wang et al., 2011), showing that transformational leaders affect how subordinates feel about their work (e.g., higher job satisfac- tion and work motivation), and how well subordinates perform their work (e.g., higher in-role and extra-role performance). Most of these studies focus on between-person differences in transfor- mational leadership, capturing transformational leadership as a “style” or behavior in general (i.e., trait transformational leader- ship measured at a single time point), which have important implications for the selection of effective leaders. Yet, no conclu- sions can be drawn about leadership development from these studies, because it is assumed that leaders are either transforma- tional, or not. To advance our understanding of transformational leadership, we focus on within-person fluctuations in leadership, acknowledging that leaders may sometimes be unable to provide their subordinates with inspiration and personalized support (see also Breevaart et al., 2014; Breevaart, Bakker, Demerouti, & Derks, 2015). That is, we focus on fluctuations in transformational leadership behavior rather than on trait transformational leader- ship. We propose that there is good reason to believe that transfor- mational leadership behavior varies from day to day. For one, leaders are not always physically present at the workplace due to the increased flexibility in working hours and work spaces (e.g., working from home; Baane, Houtkamp, & Knotter, 2010), which is known to neutralize the positive impact of transformational leadership behavior on subordinates’ performance (e.g., Howell, Neufeld, & Avolio, 2005). In addition, even if leaders are around, there may be days on which they withdraw themselves from the workplace, for example because of a bad night’s sleep, family worries, and/or a bad mood (Barnes, 2012; Hammer, Bauer, & Grandey, 2003; Podsakoff, LePine, & LePine, 2007; Scott & Barnes, 2011). Moreover, leaders may adjust their behavior ac- cording to the needs of their subordinates. That is, when subordi- nates work on a task they are actively engaged in and feel self- efficacious about, there is less urgency to motivate subordinates to perform their work (Dvir & Shamir, 2003). Adopting a within-person approach to leadership, a handful of diary studies provide support for the dynamic nature of transfor- mational leadership. For example, in a study among consultants, Tims, Bakker, and Xanthopoulou (2011) showed that subordinates were more engaged in their work on the days that their leader showed more transformational leadership behavior, because sub- ordinates were more optimistic on these days. In a similar vein, Breevaart et al. (2014) showed that naval cadets were more en- gaged in their work on the days their leaders used more transfor- mational leadership behavior, and rewarded good performance (i.e., contingent reward). The reason for this was that transforma- tional leaders provided their followers with more job resources. Moreover, the latter study showed that cadets in a leadership position used both transformational and transactional leadership behavior within the same day, and that most variance in both types of leadership could be explained by within-person differences. Following this dynamic view on leadership, in the present study, we examine on which days transformational leadership behavior should be used to sustain and/or increase subordinates’ work engagement. Job Demands–Resources Theory According to JD-R theory, the degree to which employees are energetic and enthusiastic about their work, and highly concen- trated on their work (i.e., engaged), is greatly affected by the work environment (Bakker & Demerouti, 2014; Demerouti et al., 2001). Specifically, JD-R theory distinguishes two categories of job char- acteristics; job resources and job demands. Job resources are aspects of the job that stimulate personal growth and accomplish- ment and initiate a motivational process. For example, opportuni- ties for development and performance feedback may fulfill em- ployees’ need for competence, and consequently, their willingness to invest themselves in their work role (Kahn, 1990). Job demands are proposed to initiate an energy depletion process, consuming energetic resources, which may result in job strain and health complaints. For example, constantly trying to deal with conflicting standards may wear out employees, decreasing their willingness and ability to invest themselves in their work role. Whereas it is well established that job resources like autonomy and opportunities for development contribute to employees’ en- gagement (for a meta-analysis, see Halbesleben, 2010), the rela- tionship between job demands and employee engagement is less clear. A possible explanation for the inconsistent findings regard- ing the link between job demands and employees’ feelings of engagement in their work can be found in the challenge stressor– hindrance stressor framework that was first proposed by Ca- vanaugh et al. (2000). Accordingly, all job demands cost energy, but some demands hinder personal development and goal achieve- ment (i.e., hindrance demands), whereas other demands create opportunities for personal growth and achievement (i.e., challenge demands). Hindrance demands such as role-conflict, role ambigu- ity, and hassles are unnecessary obstacles toward goal achievement and personal learning that demotivate employees. In contrast, challenge demands such as workload and time pressure result in a sense of accomplishment when they are overcome. T hi s do cu m en t is co py ri gh te d by th e A m er ic an P sy ch ol og ic al A ss oc ia ti on or on e of it s al li ed pu bl is he rs . T hi s ar ti cl e is in te nd ed so le ly fo r th e pe rs on al us e of th e in di vi du al us er an d is no t to be di ss em in at ed br oa dl y. 339DAILY JOB DEMANDS AND LEADERSHIP In support of this challenge stressor– hindrance stressor distinc- tion, Crawford, LePine, and Rich (2010) meta-analytically showed that challenge demands (e.g., workload and time urgency) and job resources (e.g., autonomy and feedback) were positively related to employee engagement, whereas hindrance demands (e.g., admin- istrative hassles and role-conflict) were negatively related to em- ployee engagement. Tadić, Bakker, and Oerlemans (2013) found similar results in their daily diary study among teachers. Specifi- cally, they showed that teachers were more engaged in their work on days that they had more challenging demands, because they experienced more self-concordant work motivation on these days. The opposite was true for hindrance demands, that is, teachers experienced less self-concordant work motivation on days they were confronted with more hindrance demands, and consequently, teachers were less engaged in their work on these days. In the present study, we focus on clear examples of challenge (i.e., workload, cognitive demands) and hindrance (i.e., role-conflict, family to work conflict) demands that have been shown to either contribute or be detrimental to employee motivation and perfor- mance (Halbesleben, 2010; LePine et al., 2005) and that are relevant to our sample of teachers. Boosting Effect Besides the main effects of resources and demands on employee well-being, JD-R theory also proposes two interaction effects. The first interaction effect is known as the boosting effect: job re- sources particularly boost employee engagement when challenging job demands are high (Bakker & Demerouti, 2014). It seems likely that employees feel especially engaged in their work on days when they have a sufficient amount of resources available to deal with challenging job demands. For example, employees who receive performance feedback from their supervisor when they work under high pressure may learn to be more efficient and effective, con- tributing to employees’ feelings of competence. Investigating the role of personal resources such as optimism and self-efficacy, Bakker and Sanz-Vergel (2013) showed that nurses were more work engaged in the weeks that both emotional demands and personal resources were high. Tadic, Bakker, and Oerlemans (2015) found a similar pattern in their daily dairy study among schoolteachers, showing that challenging demands (e.g., workload and time urgency) were more positively related to employees’ positive affect and work engagement when job resources such as social support from colleagues and performance feedback were high (vs. low). Building on this knowledge, we examine the interactions between daily challenge demands (i.e., workload and cognitive demands) and transformational leadership in predicting employees’ daily engagement. Furthermore, rather than looking at the broad categories of challenge (and hindrance) demands (Tadic et al., 2015), in the present study we look at the interaction effect between each specific demand and transformational leadership. Cognitive demands refer to the degree to which the job requires employees to be highly concentrated on their work, whereas work- load means that employees have a lot of work to do and have to work hard to finish their tasks (Bakker, Demerouti, Taris, Schaufeli, & Schreurs, 2003). We expect that even though days with high cognitive demands and workload cost energy, leaders may help to deal with these demands and thereby stimulate sub- ordinates’ feelings of engagement, using transformational leader- ship behavior. That is, transformational leadership behavior such as communicating plans for the future creates a sense of meaning- fulness regarding the work that subordinates have to perform (Arnold, Turner, Barling, Kelloway, & McKee, 2007). Also, chal- lenging employees on an intellectual level by encouraging them to think about solutions for the problems they face in their work increases subordinates’ self-efficacy (e.g., Nielsen, Yarker, Ran- dall, & Munir, 2009; Pillai & Williams, 2004). These behaviors make it likely that demands such as cognitive load and workload are viewed as opportunities to learn and demonstrate competence (i.e., become challenging). Accordingly, our first hypothesis states: Hypothesis 1: Daily transformational leadership behavior moderates the relationship between (a) cognitive demands and (b) workload on the one hand and employee work engagement on the other hand. That is, the relationship between challenge demands (i.e., cognitive demands and workload) and em- ployee engagement is stronger on the days that transforma- tional leadership behavior is high (vs. low). Buffering Effect The second interaction effect proposed by JD-R theory is known as the buffering effect. Accordingly, job resources protect employ- ees from the negative effect of hindering job demands. The reason for this is that job resources replenish energetic resources that are lost when meeting job demands, by providing employees with tools to cope with stressors at work (Bakker & Demerouti, 2014). In a daily diary study among schoolteachers, Tadic et al. (2015) found support for the buffering effect of job resources. Specifi- cally, they found that teachers experienced less positive affect and feelings of engagement on days that hindrance demands (i.e., excessive bureaucracy, role ambiguity, role-conflict, and hassles) were high, but this negative effect was reduced on the days that job resources were high (vs. low). Building on the latter study, we examine the relationship between daily hindrance demands (i.e., role-conflict and family to work conflict) and employees’ daily work engagement as moderated by daily transformational leader- ship. Role-conflict means that employees have to deal with conflict- ing assignments and/or rules and family to work conflict (FWC) means that employees have a hard time focusing on their work because of family responsibilities/problems (Bakker et al., 2003). We argue that daily transformational leadership behavior acts as a valuable resource that protects against the negative impact of these hindrance demands on subordinates’ work engagement. That is, on days that subordinates experience conflict between different roles they have to fulfill or are occupied with worries about the family, they are unable to fully concentrate on their work and waste valuable resources such as time and effort thinking about issues that may undermine their motivation to perform their work. Lead- ers who provide individualized support to subordinates on these days and focus subordinates’ attention on the positive side of their work, may buffer the detrimental influence of these hindrance demands on subordinates’ feelings of vigor, dedication, and ab- sorption (i.e., engagement). Furthermore, it has been shown that employees have access to more resources such as autonomy and social support on the days when their leader shows more transfor- mational leadership behavior (Breevaart et al., 2014), which pro- T hi s do cu m en t is co py ri gh te d by th e A m er ic an P sy ch ol og ic al A ss oc ia ti on or on e of it s al li ed pu bl is he rs . T hi s ar ti cl e is in te nd ed so le ly fo r th e pe rs on al us e of th e in di vi du al us er an d is no t to be di ss em in at ed br oa dl y. 340 BREEVAART AND BAKKER vides employees with the necessary tools to deal with the demand- ing aspects of their work. Hence, we hypothesize that daily transformational leadership behavior moderates the relationship between daily job demands and subordinates’ daily work engage- ment: Hypothesis 2: Daily transformational leadership behavior moderates the relationship between (a) role-conflict and (b) family to work conflict on the one hand and employee work engagement on the other hand. That is, the relationship be- tween hindrance demands (i.e., role-conflict and family to work conflict) and employee work engagement is weaker on the days that transformational leadership behavior is high (vs. low). Please see Figure 1 for an overview of the proposed interactions between daily challenge/hindrance demands and transformational leadership behavior. Method Participants and Procedure Participants were elementary schoolteachers from The Nether- lands who filled out an online questionnaire about their leaders’ (school principals’) behavior, their own level of work engagement, and the perceived job demands at the end of each workday for a period of 10 days. One of our reviewers expressed his or her concerns about the relevance of our research question on leadership to our particular sample. To ensure relevance, we collaborated closely with the HR department as well as the school principals and teachers before we conducted the study. Teachers spend most of their work hours on teaching, preparing classes, and meeting and interacting with parents. Furthermore, teachers indicated that they communicate daily with their school’s principal, mainly through face-to-face interactions. The school principal can show transformational leadership behavior by, for example, listening to the problems teachers face on a specific day (e.g., talking to parents about their child misbehaving in class), asking teachers what aspects of their work they enjoy (i.e., enthuse them about their work tasks), and informing teachers about opportu- nities to develop themselves further (e.g., suggesting to participate in a time-management workshop when they have difficulties managing their tasks). In the current study, 42.9% of the variance in trans- formational leadership was explained at the day level, indicating that the extent to which the same school principal uses transfor- mational leadership varies greatly from day to day. Teachers were informed about the study through their organi- zation and received an e-mail from the first author with the invitation to the online questionnaire at the end of every day. On the first day, participants were requested to fill out some additional questions about demographics, after which they received the same short questionnaire on every succeeding day. To ensure anonymity and confidentiality and still be able to identify the questionnaires filled out by the same person on different days, we created panels in Qualtrics containing the email address and associated code (i.e., 1, 2, 3, etc.) of all participants. A few weeks after the data was collected, participants received a report about the results of the study. In total, 1109 teachers received an invitation to participate in the study, of which 585 teachers (53%) filled out at least one of the questionnaires. Because we were interested in within-person fluc- tuations in our study variables, we kindly requested our partici- pants to fill out the questionnaires for at least five days (i.e., regular working week). Accordingly, we removed 314 participants who filled out fewer than five questionnaires from further analy- ses, resulting in a final sample of 271 teachers (24% of the original sample that was approached) who filled out the questionnaires for 5.68 days on average. The total number of data points is consid- erable, namely 271 � 5.68 days � 1539 observations. To ensure that teachers were not forced to answer questions about their principal’s behavior on the days that they did not interact, we asked teachers to indicate whether they had interacted with their principal. If they had not interacted with their principal on a specific day, they did not receive any questions about their prin- cipal’s behavior. On average, teacher interacted with their princi- pal on 5.68 days out of the 7.31 days that they filled out the questionnaire. The sample includes 219 women (80.8%) and 52 men (19.2%), with a mean age of 46.25 (SD � 11.25), ranging from 24 to 63 years. Most participants were married or cohabiting (80.8%), finished higher vocational training (79.7%), and had a permanent contract (99.6%). On average, participants had 22.33 (SD � 11.07) years of work experience and worked in the current organization for a period of 17.34 (SD � 10.78) years. Daily challenge demands • Cognitive demands (H1a) • Workload (H1b) Daily hindrance demands • Role-conflict (H2a) • FWC (H2b) Daily work engagementDaily transformational leadership H1a,b H2a,b Figure 1. Hypothesized interaction between daily demands and daily transformational leadership. FWC � family to work conflict. T hi s do cu m en t is co py ri gh te d by th e A m er ic an P sy ch ol og ic al A ss oc ia ti on or on e of it s al li ed pu bl is he rs . T hi s ar ti cl e is in te nd ed so le ly fo r th e pe rs on al us e of th e in di vi du al us er an d is no t to be di ss em in at ed br oa dl y. 341DAILY JOB DEMANDS AND LEADERSHIP Of the 1139 teachers invited to participate in our study, 271 teachers filled out the survey. Although this is a large sample with high statistical power (N � 271 � 5.68 days � 1539 observa- tions), the response rate is quite low (23.79%), meaning that our results could be affected by nonresponse bias. Those who partic- ipated in our study may differ from those who did not participate. To test for such a bias, we performed an attrition analysis to examine differences between those employees who participated 5 days or more (N � 271) and those who participated less than 5 days and were removed from our final sample (N � 314). There were no significant gender, age, marital status, education and work experience differences between these two groups. Yet, we did find that those who participated less than five days were on average a bit less engaged in their work (�M � .214, p � .01), perceived their leader as less transformational (�M � .221, p � .01), and had a lower quality relationship with their leader (�M � .314, p � .001). However, because we examined within-person fluctuations (i.e., fluctuations from employees’ average experiences) rather than between-person differences (i.e., employees’ overall experi- ences), it seems unlikely that our results have been influenced by nonresponse bias. That is, rather than looking at absolute differ- ences in work engagement (i.e., low/high), we look at fluctuations in work engagement from people’s baseline (i.e., lower/higher). Measures We adapted the time-frame of all questionnaires so the items refer specifically to the day, which is common practice in diary research (Ohly, Sonnentag, Niessen, & Zapf, 2010; Xanthopoulou & Bakker, 2013). Participants could answer all statements on a 7-point scale, ranging from 1 (completely disagree) to 7 (com- pletely agree). All scales showed good alpha reliabilities (please see Table 1). Daily challenging demands. Daily cognitive demands and daily workload were measured with three items each (Bakker et al., 2003). An example item is: “Today, my work required a high level of concentration” (i.e., cognitive demands) and “Today, I had to work very hard” (i.e., workload). Daily hindrance demands. Daily hindrance demands were measured with three items each. An example item of role-conflict is: “Today, I had to deal with conflicting guidelines or rules” (Rizzo, House, & Lirtzman, 1970). An example item of strain- based family to work conflict is: “Today, I did not enjoy my job because I worried about my home situation” (Geurts et al., 2005). Day-level transformational leadership. Day-level transfor- mational leadership was measured with four different scales from the Transformational Leadership Inventory (TLI; Podsakoff, MacKenzie, Moorman, & Fetter, 1990) that are most closely related to Bass’ (1985) conceptualization of transformational lead- ership. Example items are: “Today, my leader inspired me with his/her plans for the future” (articulating vision; 3 …