Leading the Unsatisfied Generation…
How Do Intrinsic and Extrinsic Work Values Mediate the Relationship Between
Millennial Identification and Job Satisfaction and to what Extent Do
Statement of originality
This document is written by Natalie de Vries who declares to take full responsibility for its content. I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
Amsterdam, 29.06.2018
Table of content
List of tables and figures ... 4
Abstract ... 6
Introduction ... 7
2. Literature review ... 11
2.1 Millennials ... 11
2.1.1 Defining Millennials as a generation... 11
2.1.2 Millennial identification ... 14
2.1.3 Who are these Millennials? ... 15
2.1.4 Millennial identification and job satisfaction ... 16
2.2 Work values ... 18
2.2.1 Extrinsic work values... 19
2.2.2 Intrinsic work values ... 21
2.3 Leadership ... 24 2.3.1 Transformational leadership ... 26 2.3.2 Transactional leadership ... 28 3. Method ... 30 3.1 Research design ... 30 3.2 Sample ... 33 3.3. Measurements ... 34 3.3.1 Translation... 34 3.3.2 Scale reliability ... 35 4. Results ... 40 4.1 Preliminary steps ... 40 4.2 Direct effects ... 44 4.3 Mediation effects ... 45
4.4 Moderated mediation effect ... 50
4.4.1 Simple moderation analysis ... 50
4.4.2 Exploratory analyses ... 55 5. Discussion ... 64 5.1 Summary... 64 5.2 Theoretical implications ... 65 5.3 Limitations ... 71 5.4 Future research ... 72 5.5 Practical implications ... 75 5.6 Conclusion ... 76 6. Reference list ... 78 Appendix ... 94 Survey in Dutch ... 94
List of tables and figures
Tables
Table 1 Means, standard deviations and correlations of 43
mean variables and control variables
Table 2 Hierarchical regression of job satisfaction 45
Table 3 Simple Mediation analysis of Millennial Identification 48
via Extrinsic work values on job satisfaction
Table 4 Simple Mediation analysis of Millennial identification 49
via Intrinsic work values on job satisfaction
Table 5 Simple Moderation analysis of Millennial identification 53
on extrinsic work values for transformational leadership
Table 6 Simple Moderation analysis of Millennial identification 53
on intrinsic work values for transformational leadership
Table 7 Simple Moderation analysis of Millennial identification 54
on extrinsic work values for transactional leadership
Table 8 Simple moderation analysis of Millennial identification 54
on intrinsic work values for transactional leadership
Table 9 Hierarchical regression of Millennial identification 56
Table 10 Hierarchical regression of intrinsic work values 57
Table 11 Hierarchical regression of extrinsic work values 57
Table 12 Hierarchical regression of job satisfaction 57
Table 13 Simple moderation analysis of intrinsic work values 60
on job satisfaction for transformational leadership
Table 14 Simple moderation analysis of intrinsic work values 60
on job satisfaction for transactional leadership
Table 15 Simple moderation analysis of extrinsic work values 61
Table 16 Simple moderation analysis of extrinsic work values 62
on job satisfaction for transactional leadership
Table 17 Simple moderation analysis of Millennial identification 62
on job satisfaction for transformational leadership
Table 18 Simple moderation analysis of Millennial identification 63
on job satisfaction for transactional leadership
Figures
Figure 1 Research model 31
Figure 2 Interaction effect of intrinsic work values on job 61
Abstract
This study aimed to examine the mediating effects of work values on the relationship of Millennials and job satisfaction. While a lot of research on Millennials has been done, no research has ever used a theoretical approach to categorize individuals in this generation. Therefore this study provides a new theoretical measurement method based on
generational identification, in which individuals are asked to which extent they identify with the generation Millennials. In addition, both transformational and transactional leadership were added as moderators in the mediated relationship of Millennial identification through work values on job satisfaction.
This research is based on data collected of 149 Dutch employees through a questionnaire. Results reveal that employees with high Millennial identification have both high intrinsic as well as high extrinsic work values, which was a remarkable outcome since it was always assumed that both work values were antagonistic. Furthermore the results of exploratory analyses disclosed that when the traditional measuring method -categorizing Millennials based on birth cohorts- was executed none of the relations were found to be significant. In conclusion, this study contributes to literature by introducing the concept of synergy in both work values amongst Millennials and by introducing a new relevant measuring method for generational research.
Key words: Millennials, Millennial identification, generational identity, intrinsic work values,
Introduction
With 78 million people born between the years 1980 and 2000, Millennials are the biggest cohort generation of all times (Rainer & Rainer, 2011) and within two years’ time will make up for one third of the global workforce (AARP & Murphy, 2007). With extreme high levels of narcissism -the highest found in thirty years-, self-confidence and self-absorption this
generation has earned themselves a new name: Generation Me (Twenge, Konrath, Foster, Campbell, & Bushman, 2008a, 2008 b). Being characterized as a generation lacking in loyalty and work ethics, they clearly differ from their precedent generations (Marston, 2007). These differences have led to a psychological battlefield on the work floor, “wherein buttoned-down, self-centered Millennials clash with their stodgy, rule-abiding Baby Boomer bosses” (Kowske, Rasch, & Wiley, 2010, p. 265). Millennials are excellent in sending contradictory signals, having extremely high career expectations but unwilling to sacrifice their work-life balance by working extremely hard. It leaves managers wondering is it just their perception or are these youngsters really different (Smola & Sutton, 2002)?
The growing interest in this specific generation can be noticed in the amount of attention it has received in management magazines, books, business consultancy publications and even national newspapers (Parry & Urwin, 2011). Also research has grown rapidly in the last 15 years trying to explain these Millennials. Nevertheless, research has taken a non-theoretical approach, examining differences between generations based on birth years, but the precise definition of these cohorts differs amongst all (Parry & Urwin, 2011). This has led to
contradictory results, raising questions about the legitimacy of the differences between generations on the work floor (Costanza & Finkelstein, 2015) and suitable methods to
examine it (Parry & Urwin, 2011). Therefore generational research is in need of a new measurement method with a deeper understanding of the generation as an individual and social construct (Lyons & Schweitzer, 2017; Lyons, Urick, Kuron, & Schweitzer, 2015; Parry & Urwin, 2011). This study will operationalize a measurement method based on generational identity theory. Millennial identification measures the degree to which an individual identifies him- or herself with the norms, values and collective memories of Millennials (Lyons & Kuron, 2014).
Despite its limitations due to the birth cohort perspective, over the last few years
evidence has grown rapidly that Millennial workers do have significantly different values, attitudes and expectations in work than the current generations (Lyons & Kuron, 2014; Lyons & Schweitzer, 2017; Twenge, Campbell, Hoffman, & Lance, 2010). Furthermore, Millennials also differ in personality, which has important consequences at work, as personality influences turnover, satisfaction and leadership (George, 1992). Therefore it is interesting to examine the levels of intrinsic and extrinsic work values of employees that identify themselves as Millennials, instead of assuming homogeneity based on their date of birth.
Knowing the relation between work values and Millennial identification is important
in understanding how organizations can influence their satisfaction levels, since Millennials are a generation with increased levels of depression and anxiety (Twenge, 2006). Research has proven that dissatisfied employees are less committed (Lok & Crawford, 2004), have a lower productivity and are even more likely to quit their jobs (Hackett & Guion, 1985; Hulin, Roznowski, & Hachiya, 1985; Kohler & Mathieu, 1993). As a response, this study explores the effect work values have on job satisfaction levels of employees with high Millennial
identification, in order to provide guidance for organizations on how to increase these satisfaction levels.
In line with these findings, several authors have already stated that this self-confident
and dissatisfied generation is in great need of strong leadership (Gursoy, Chi, & Karadag, 2013). However, little research has been done on how leadership influences Millennial’s work values and different authors call out for further investigation (Gursoy et al., 2013; Ng, Schweitzer, & Lyons, 2010). Therefore, transformational and transactional leadership will be studied separately as moderators on the mediated relationship between Millennial
identification through both intrinsic and extrinsic work values on job satisfaction. As the two leadership styles differ significantly, it’s expected that they relate differently to both intrinsic and extrinsic work values. As a result it’s proposed that both leadership styles will have opposite moderating effects on Millennial identification through the same work value on job satisfaction. So for example, it’s expected that transformational leadership strengthens the mediated relationship of Millennial identification through intrinsic work values on job satisfaction and that transactional leadership will weaken the mediated relationship.
It’s fundamental to gain a deeper understanding of the values in the work place of this generation; however, measured correctly. This research contributes to social science by introducing a new relevant method to measure the extent to which an individual identifies themselves as a Millennial, instead of automatically assigning them to this generation based on their birth years. Second, practical relevance is also established as this study aims to understand the work values of employees with high Millennial identification and explore their effect on job satisfaction. Moreover, adding the two most discussed leadership styles -transactional and transformational leadership- as moderators, will enhance the theoretical
and practical recommendations for managing this high demanding generation. In conclusion, this study will offer a new view on generational research striving to give a deeper
understanding of the real Millennials, as they are just too big to ignore.
In the next section the relevant literature for this study will be extensively reviewed, in which all concepts will be explained and the hypothesized relationships introduced. Second the research model, chosen research design and methods will be described. Finally the results of the hypothesized relations will be presented, followed by a discussion of the results.
2. Literature review
2.1 MillennialsGeneration Me, Nexters, Baby Boom Echoers, Generation Y, the Nexus Generation and Millennials, this generation has been given many different names trying to reflect this unique group (E. Ng et al., 2010). In this study the term used will be the term most commonly referred to: Millennials. Almost all research have labeled an individual as a Millennial based solely on their birth date. Even in this categorization great differences in birth categories exist, but the majority range between the years 1980 and 2000.
2.1.1 Defining Millennials as a generation
A generational group is identified by birth years and share momentous events at crucial stages in their development (Kupperschmidt, 2000). Although they display differences, they grew up in the same time period sharing the same life experiences, which cause them to develop similar attitudes and beliefs (Beldona, Nusair, & Demicco, 2009; Kupperschmidt, 2000; Meriac, Woehr, & Banister, 2010). This is why most studies have based their
generational classification on birth cohorts. In the last 15 years there has been a tremendous growth in research on the generation at the work place phenomenon, however there is minimal empirical evidence supporting all the generational differences (Costanza & Finkelstein, 2015). Additionally, or probably because of the lack of evidence, a lot of the results are contradictory, raising questions about the legitimacy of generational differences (Lyons & Schweitzer, 2017). The problem with existing research lays in the categorization of Millennials based on these birth cohorts, which has recently gained a lot of criticism
(Twenge, Campbell, Hoffman, & Lance, 2010).
generations as cohorts, it suggests that they are homogenous groups with fixed, measurable and observable behavioral commonalities based on birth year cut-offs (Ryder, 1965). Or as Karl Mannheim “the father of modern generational theory” in 1952 already noted, “the sociological phenomenon of generations is ultimately based on the biological rhythm of birth and death, but to be based on a factor does not necessarily mean to be deducible from it, or to be implied in it” (p. 292).
Second, the boundaries for a birth cohort are unclear, the start and ending years are
used inconsistently. One research could identify a Millennial based on their birth cohort whereas another would identify them as Generation Z (Wang & Peng, 2015). Age is seen as a biological fact but generation is mostly seen as a social construct and with any social
construct boundaries are discussed and change over time and circumstances (Campbell, Campbell, Siedor, & Twenge, 2015; Lyons & Schweitzer, 2017; Nakai, 2015).
Third, as mentioned a generation shares the same historical life events, leading to
similarity in attitudes and beliefs. This has the strongest influence during childhood and young adolescence. Generations are categorized based on these birth cohorts, however research suggests that these cohort effects are linear with change happening steadily over time, rather than abrupt changes at birth year cutoffs (Jean M. Twenge, Konrath, Foster, Keith Campbell, et al., 2008; Jean M. Twenge, Zhang, & Im, 2004). Additionally, this
generational category of birth years could also lead to falsely categorizing someone whose birth year is just between two different generations (Wang & Peng, 2015).
The fourth flaw is the age related effect, any generational difference might be related
to age, so as Baby boomers are currently older they might have different values than the young Millennial generation solely due to the fact that they have matured more (Dencker, Joshi, & Martocchio, 2008). In 1952 Karl Mannheim already argued that to apply generations
as a useful construct it should be used above and beyond age and period (Lyons, Urick, Kuron, & Schweitzer, 2015, p. 348).
So viewing generations as birth cohorts is far too simplistic and overlooks the
theoretical underpinning of the generation as a social construct and presumes homogeneity in age cohorts (Lyons & Kuron, 2014).
Although birth cohorts are easily measurable, it is inadequate as means of operationalizing generations in research and therefore use of a stronger concept is needed (Lyons &
Schweitzer, 2017). The solution could lie in generational identification, which can be explained through the concept of social identity theory. This theory claims that people identify with social categories (such as generations), in order to improve their self-concept and define themselves respective to others in their social environment (Ashforth & Mael, 1989).
In line with this, Lyons & Schweitzer (2017) found that employees develop social
identities related to their work and therefore do make use of a generation as a social category. By using identification it is possible to examine to what degree people identify themselves with collective generational memories, values and norms instead of assuming homogeneity because of commonality in birth years (Lyons & Kuron, 2014).
To date, generational identification has not been researched a lot, more specifically
only a recent study by Lyons & Schweitzer (2017) studied the vitality of this construct in general. So although Millennial identification has never been operationalized in a study before, it still is a much stronger measurement method, as it assigns only the individuals that identify with Millennials to this generation instead of automatically assigning them based on
their birth cohorts. Moreover, it removes the flaws tied to the objective measure of birth cohorts and adheres to the definition of a generation as a social construct. Therefore
Millennial identification will be used, in which a Millennial is determined based on the extent to which an individual actually sees themselves as a Millennial.
2.1.2 Millennial identification
As generational identity theory is relatively new, only a few authors have studied the definition on how a generation is shaped on the basis of individual identities (Kelan, 2014). Joshi et al., (2011) included both the cognitive and emotional aspect in the definition of generational identity, as they suggest that people identify themselves on the basis of their knowledge of that generation and place great importance on group identification as they invest emotionally in the group construct. Referring back to social identity theory, it acknowledges that when people relate to a group they see themselves as one in their perceptions, attitudes and beliefs and perceive people from outside the group as different (Hogg & Terry, 2000). This phenomenon especially comes to life in social situations, such as in a work environment, which could explain why forms of generational clashes on the work floor exist.
Basing generations on identification has several advantages. First of all it allows
people to decide to what generation they relate themselves and more importantly to what degree. By measuring the degree of identification, generation is treated as a continuous variable instead of treating generation as a discrete category where your birth cohort decides if you belong to the generation or not (Lyons & Kuron, 2014). Second, it gives individuals the possibility to identify with multiple generations in varying degrees. Third, it refrains from the confusion between age, cohort and period and also applies age and cohort
as inseparable influences on identity (Lyons & Kuron, 2014). Lastly, it acknowledges the influence of generational culture field, which are thoughts and actions that emerge “at a specific point in history that may have adherents from multiple birth cohorts, but is likely centered around a core of a certain age range” (Lyons & Schweitzer, 2017, p. 210).
2.1.3 Who are these Millennials?
Millennials grew up during the “empowerment years in which everyone won and everyone got a medal” (Gursoy et al., 2013, p. 42) and where participating is rewarded instead of performance. This “praise for anything” attitude leads to growing ego’s, higher levels of self-esteem and an extreme need for constant praise (Hill, 2002; J. Twenge, 2013; Jean M.
Twenge, 2010). Born in a 24/7 online world where they are used to “I want it all” and “I want it now”, raised with the belief that they can become anything, while being nurtured by their parents has led to them being portrayed as a spoiled and entitled generation (Howe & Strauss, 2000; J.M. Twenge, 2006).
Indeed, research has argued that Millennials in the Netherlands (Smits, Dolan, Vorst,
Wicherts, & Timmerman, 2011) have higher levels of narcissism, neuroticism and self-confidence than their preceding generations (Lyons & Kuron, 2014). This change in
personality has a significant impact on their career and work perspective, as they expect a promotion within the first 15 months of their first job and demand their salary to increase with 63 per cent in five years’ time -which is 5 times the average increase- (E. Ng et al., 2010). Of course they expect their employers to accommodate all their expectations, but on the other hand want a perfect balance between their work and social life (Gursoy et al., 2013). So they are indeed a high demanding generation full of confidence, with great expectations.
2.1.4 Millennial identification and job satisfaction
Before examining the relationship between Millennial identification and job satisfaction it is essential to briefly discuss the definition and consequences of job satisfaction.
Locke’s definition of job satisfaction is the most common in research, defining it as a “…pleasurable or positive emotional state resulting from the appraisal of one’s job or job experiences” (Locke, 1976, p. 1304). In this definition both feeling and thinking are
recognized, which implies that when we evaluate our job we include both affective (feelings) and cognitive (thinking). This response can be about the job in general or specific aspects of the job. Job satisfaction can have numerous consequences for organizations, as it is a determinant of organizational performance, (Angle & Perry, 1981; Riketta, 2002) as well as individual performance (Judge, Thoresen, Bono, & Patton, 2001; Wiggins & Moody, 1983). Furthermore, research has proven that satisfaction is associated with client outcomes, (Buffum & Konick, 1982; Schwartz & Will, 1961) organizational effectiveness (Laschinger, 2001; Miller, 1978) and organizational citizenship behaviors (Organ & Ryan, 1995). On the negative side, dissatisfied employees are less committed (Lok & Crawford, 2004), more likely to be absent and quit their jobs (Hackett & Guion, 1985; Hulin et al., 1985; Kohler &
Mathieu, 1993).
Looking more closely at job satisfaction levels between generations, both Westerman
& Yamamura (2007) and Cennamo & Gardner (2008) found no significant evidence that generations differ in job satisfaction levels amongst each other. While on the contrary, Costanza, Badger, Fraser, Severt, & Gade (2012) compared over 20 studies in their meta-analysis and concluded that Millennials were less satisfied with their jobs compared to Baby Boomers and Generation X. Moreover, they even found that Millennials were more likely to leave their jobs than older generations. This difference could be related to the fact that birth
cohort was used to identify generations, since research has found that age and job satisfaction are chronically related, as satisfaction increases when age increases (Hunt & Saul, 1975; Kacmar & Ferris, 1989, 1989; T. Ng & Feldman, 2010). Therefore it is interesting to find out based on their Millennial identification if employees also experience lower job satisfaction levels.
When controlling for age and using Millennial identification, it is still expected that
employees with high Millennial identification have a negative relation with job satisfaction, in line with findings of Costanza et al., (2012). The negative relationship is expected for two important reasons. First, Millennials have high career expectations which differ greatly from reality (Dries, Pepermans, & Kerpel, 2008) and this discrepancy can lead to disappointment influencing their job satisfaction levels negatively. Second, Millennials are a generation that in general have lower well-being levels, as Twenge (2006) found that Millennials have increased levels of anxiety and depression, which can also negatively influence the level of job satisfaction. So when applying Millennial identification it is expected that the more an individual identifies themselves as a Millennial, the more they behave like a Millennial is said to behave, resulting for instance in lower job satisfaction levels. Therefore the expectation arises that the higher the levels of Millennial identification the lower the levels of job
satisfaction. Leading to the first hypothesis in which it is proposed that employees with high Millennial identification have low job satisfaction levels.
H1: Millennial identification is negatively related to job satisfaction
So it is assumed that generations differ in norms and values. These norms and values
are the meanings employees attribute to their work, which among other things mediate job satisfaction (Goldthorpe, Lockwood, Bechhofer, & Platt, 1968; Kalleberg, 1977). In order to
know the effect on job satisfaction it is important to look at intermediate values. Not only to explain satisfaction, but also to test if these attributed characteristics of Millennials are indeed correctly assigned. If it is indeed true that certain values of Millennials have shifted from their previous generations, this will have many implications in the work field, for example it can impact the effectiveness of certain leadership styles (George, 1992).
First, the intrinsic and extrinsic work values of Millennial identities are studied,
followed by the mediated relationship with job satisfaction. Finally, the influence of leadership on the mediated relationship will be examined.
2.2 Work values
Work values are the perceptions of an individual about the importance of specific outcomes connected to the attributes of work (Elizur, 1984). Work values influence the way people feel about their work (Spence, 1985), their attitudes and behaviors (Dose, 1997; England, 1967). Moreover, it also influences their job decisions, (Judge & Bretz, 1991) commitment and job satisfaction (Butler & Vodanovich, 1992; Cheung & Scherling, 1999; Drummond & Stoddard, 1991; Locke, 1976). As work values influence so many elements, it becomes essential for leaders to identify and understand the work values of Millennials in order to positively influence them.
The two most common distinctions are intrinsic and extrinsic work values. Intrinsic
work values relate to the desire of employees for self-development and achievement, the amount of influence over their work, creating relationships and the meaningfulness of work (Knoop, 1994). An individual that has intrinsic work values places great importance on a job that “provides variety and responsibility, offers a challenge, enables the employee to see the results of what he or she does, and has a significant impact on others” (Twenge et al., 2010,
p. 1124). While intrinsic work values are more related to the process of the work, extrinsic work values are related to the consequences of work (Elizur, 1984).
2.2.1 Extrinsic work values
Extrinsic work values are associated with the achievement of expectations which are based on instrumental resources detached from the meaning of work, such as status, financial success and prestige (Marini, Fan, Finley, & Beutel, 1996; Watts, 1992). Millennials express high career expectations mainly in forms of extrinsic components such as a high salary, promotion, status, etc. They expect immediate recognition from their employers through extrinsic rewards such as a title or a salary increase (Gursoy et al., 2013).
Although Gursoy’s (2013) findings lead to the suspicion that Millennials have high
extrinsic work values, different studies are still showing mixed findings. Hansen & Leuty, (2012) found that this generation placed less importance on status, seen as an important element of extrinsic work vales, while contrarily Krahn & Galambos (2014) and Twenge et al., (2010) both found support that Millennials do place a greater value on extrinsic rewards. The differences could be due to the design of the study. In general, but most definitely in
generational studies a longitudinal study in which data collection starts at a young age and follows the participants through their (work) lives is most reliable (Schaie, 1965). Therefore greater importance is placed on the findings of Krahn & Galambos (2014) and Twenge et al., (2010) since they consitute longiditudinal studies. Based on these results it’s expected that employees with high Millennial identification place greater importance on extrinsic work values.
Since it’s expected that Millennials have a high relation with extrinsic work values, it is interesting to see what effect this could have on job satisfaction, does it increase or decrease the work experience? The answer to this question lies within self-determination theory (Deci & Ryan, 1985). Self-determination theory is an approach to human motivation and personality and hypothesizes that individuals have three universal psychological needs, namely belonging to autonomy (their behavior is freely chosen), competence (feeling effective and skillful in their activities) and relatedness (care for others and feeling cared) (Deci & Ryan, 2000). According to self-determination theory the satisfaction of these
psychological needs is essential for optimal functioning, psychological growth and well-being (Ryan & Deci, 2000). However, when these needs are not satisfied it will result in negative psychological consequences (Van Steenkiste et al., 2007).
As mentioned, extrinsic work values include material aspects of work, such as pay,
benefits, and job security. As they are so extrinsically oriented towards social approval, they disregard their own personal needs and interests (Boven & Gilovich, 2003; Fromm, 1976). Moreover, self-determination theory explains that the pursuit of extrinsic values, even when accomplished, are less likely to generate job satisfaction. This is because extrinsic value pursuit tends to be contradictory with the basic need satisfaction (autonomy, relatedness and competence) (Van Steenkiste et al., 2007). This unfulfillment of the basic psychological needs results in negative individual outcomes, such as lower self-esteem, social productivity and is related to lower physical health (Kasser & Ryan, 1993, 1996; McHoskey, 1999; Van Steenkiste, Duriez, Simons, & Soenens, 2006). Consistent with theory, studies using self-determination theory found that employees who placed greater importance on external values were less satisfied with their jobs (Kasser & Ryan, 2001; Van Steenkiste et al., 2006) and even tended to be less happy with their lives (Van Steenkiste et al., 2007). Leading to
the following hypothesis:
H3: extrinsic work values are negatively related to job satisfaction
Based on the expected relationships between Millennial identification, extrinsic work
values and job satisfaction, it is proposed that extrinsic work values mediate the relationship between Millennial identification and job satisfaction.
H4: extrinsic work values mediate the relationship between millennial identification and job satisfaction
2.2.2 Intrinsic work values
As individuals with extrinsic work values are motivated to achieve materialistic or extrinsic rewards, individuals with intrinsic work values are motivated to work for their own sake in forms of self-development, achieving high levels of autonomy or being involved in decision making (J. Twenge et al., 2010). Although theory clearly directs us to suspect that Millennials mostly have high extrinsic work values, some authors have also mentioned that Millennials place great value on intrinsic aspects of the work, such as training and mentoring to remain attractive in the job market (Loughlin & Barling, 2001; Lyons, 2004). Moreover, they prefer challenging jobs in which they can continuously develop their skills and have high
responsibility levels, which indicates also the importance of intrinsic work values (Walsh & Taylor, 2007).
Nevertheless, when comparing the three recent generations, Walsh and Taylor
(2007) found that Baby Boomers and Generations X’ers ranked intellectual stimulation, known as a key element of intrinsic work value, much higher than Millennials and in return Millennials ranked economic return higher. Implying that Millennials do have some form of
intrinsic work values, but place greater value on extrinsic. In relation, Cennamo & Gardner (2008) found that Millennials place more importance on status as extrinsic work value. Additionally, a longitudinal study done over 15 years by Twenge et al., (2010) demonstrated that in total Millennials valued an intrinsic rewarding job less. Leading to hypothesize that employees with high Millennial identification are negatively related to intrinsic work values.
H5: Millennial identification is negatively related to intrinsic work values
Several studies have shown that aspects of intrinsic work values influence job
satisfaction, such as autonomy (Adams & Bond, 2000). Referring back to self-determination theory, claiming that pursuing intrinsic work values facilitates satisfaction of the basic psychological needs (autonomy, competence and relatedness), which will lead to higher well-being levels (Deci & Ryan, 2000; Kasser, 2002). Additionally, intrinsic work value
orientation is identical to the need for personal growth and self-actualization at work (Deci & Ryan, 2000; Ryan & Deci, 2000). As a consequence, when there are less development
opportunities or low personal recognition, job satisfaction decreased significantly (Furlong & Glover, 1998).
On the contrary, there are also authors that couldn’t provide any proof for the relationship between intrinsic work values and satisfaction (Drummond & Stoddard, 1991; Knoop, 1994). This could be due to the limitations those studies have, with a very small sample (N=69) (Drummond & Stoddard, 1991) and a very narrow sample (only educators) (Knoop, 1994) the results become less reliable and generalizable, but most importantly these studies were cross sectional, while longitudinal studies could provide proof of the
relationship (Taris & Feij, 2001).
most powerful determinant of job satisfaction. In line with this, Elizur (1984) argued that intrinsic rewarding jobs create mental stimulation, which will lead to psychological satisfaction. This is exactly in line with the arguments of Deci & Ryan (2000) and Kasser (2002) amongst many others. Since intrinsic values such as challenge, variety and social connections relate to the inherent psychological and cognitive satisfaction of working, it’s expected that intrinsic work values are positively related to job satisfaction.
H6: intrinsic work values are positively related to job satisfaction
As it is expected that the relationships between Millennial identification and intrinsic
work values and between intrinsic work values and job satisfaction will be significant, it is proposed that intrinsic work values mediate the relationship between Millennial
identification and job satisfaction.
H7: intrinsic work values mediate the relationship between millennial identification and job satisfaction
Several studies found that when there is a discrepancy between the work values of
the employee and the workplace norms, job satisfaction levels of an employee will
significantly decrease (Mitra, Jenkins, & Gupta, 1992). On the contrary, when work values of employees were congruent with the values of their managers, both job satisfaction and commitment increased significantly (Meglino, Ravlin, & Adkins, 1989). Since most managers belong to other generations than Millennials a discrepancy between the work values of both could occur, which can indirectly lead to decreased job satisfaction levels. Therefore, it is important to understand how transformational and transactional leadership influence intrinsic and extrinsic work values of Millennials.
2.3 Leadership
Two leadership styles have dominated the literature for the past 40 years. Although a lot more leadership styles have emerged over time, transformational and transactional
leadership have been most often compared (Burns, 1978). A transformational leader values the development of a follower and encourages them to achieve their goals. They are
characterized by their ability to articulate a vision and by emotionally and intellectually supporting their followers in order to raise both leader’s and follower’s level of motivation and morality (Kuhnert & Lewis, 1987). In contrast, transactional leaders put little effort into personal values or development and are more focused on the performance of a follower. The latter is based on exchange theory, in which the leader exchanges something of value with the follower if the performance expectation is met, basically to meet their self- interests (Bass, 1999; Lowe, Kroeck, & Sivasubramaniam, 1996). On the contrary a
transformational leader motivates the follower to move beyond immediate self-interests (Bass, 1999).
Over the years several dimensions of transformational and transactional leadership have been established. Transactional leadership is determined by three dimensions; contingent reward and management by expectation, which is divided in active and passive management by exception (Judge & Piccolo, 2004). With contingent reward the leader clarifies what is expected of the follower and how the follower is rewarded for that effort (Bass, 1999). Management by exception relates to the concept of the manager taking action to correct the behavior of the follower based on the results of leader follower transactions. Lastly, active leaders follow the subordinates and take action before the behavior creates problems, while passive leaders take action after the behavior has created problems (Judge
& Piccolo, 2004). Although three dimensions are identified, research (Bass & Avolio, 1994; Lowe et al., 1996) has found support for only one dimension being effective; contingent reward.
Transformational leadership consists of four dimensions. The first is idealized
influence also known as charisma. “Charismatic leaders display conviction, take stands and appeal to followers on an emotional level”, which leads to followers identifying with these leaders, according to Judge & Piccolo (2004, p. 755). Second, inspirational motivation concerns having a clear vision that appeals to the majority of a group and inspires them. These leaders have a clear but challenging goal with high standards always directed to a better future (Judge & Piccolo, 2004). Third, intellectual stimulation contains the ability of leaders to challenge employees by taking risks and encouraging them to use their creativity (Bass, 1985). Even more extreme, transformational leaders are able to lift average people to exceptional heights by appealing to their morals (Bass & Steidlmeier, 1999). Last,
individualized consideration is concerned with leaders showing individualized attention to subordinates, taking their personal needs into consideration and showing support for their efforts (Avolio & Bass, 1995).
Looking at both these leadership theories, it is interesting to know how they influence the mediated relationship of an employee with high Millennial identification through their intrinsic and extrinsic work values, leading to certain levels of job satisfaction. For instance, it is expected that an employee with high Millennial identification has high extrinsic work values, which lead to dissatisfaction as they do not satisfy the three basic psychological needs (Kasser & Ryan, 2001; Van Steenkiste et al., 2006). Knowing that dissatisfaction has significant consequences for organizations in terms of less commitment, lower productivity, high absenteeism and high turnover it is useful to examine if
transformational or transactional leadership can influence the mediated relationship in order to increase job satisfaction levels (Hackett & Guion, 1985; Hulin et al., 1985; Kohler & Mathieu, 1993; Lok & Crawford, 2004).
2.3.1 Transformational leadership
Intrinsic work values
Transformational leaders mostly respond to intrinsic work values through stimulating self-development, by giving followers high responsibility and encouraging them to take risks (intellectual stimulation). Moreover, several studies have argued that transformational leadership emphasizes the importance of intrinsic rewards (Bass, 1985; Burns, 1978). This could be due to the fact “that transformational leadership is associated with finding a higher purpose” (Arnold, Turner, Barling, Kevin Kelloway, & Mckee, 2007, p. 195) in terms of
meaningfulness of the job, which as noted before is an element of intrinsic work values (Knoop, 1994). Additionally, transformational leaders, unlike transactional leaders, show higher levels of consideration towards their followers, which has a positive impact on the relationship between the supervisor and the follower. In return this could influence the levels of intrinsic work values of an employee because as known, employees with intrinsic work values place great importance on this social aspect of having meaningful relationships.
All these elements display that transformational leaders appeal to the intrinsic work
values of an employee. Therefore transformational leadership will influence the negative relationship between Millennial identification and intrinsic work values, such that when transformational leadership is high the negative relationship between Millennial
identification and intrinsic work values will become weaker. Second, high levels of transformational leadership in turn will strengthen the positive relationship between
intrinsic work values and job satisfaction, leading to an increase in job satisfaction levels.
So as it is proposed that the relationship between Millennial identification and job
satisfaction is mediated by intrinsic work values, it is expected that transformational leaders strengthen this mediated relationship, in such a way that the negative relation between Millennial identification and intrinsic work values will become weaker, when employees perceive high transformational leadership versus low. In turn, increased intrinsic work values will lead to higher job satisfaction levels.
H8: Transformational leadership moderates the mediated relationship between Millennial identification and job satisfaction through intrinsic work values, such that when
transformational leadership is high (vs. low) Millennial identification will have a less negative effect on intrinsic work values, which in turn will positively relate to job satisfaction.
Extrinsic work values
Transformational leadership is closely related with intrinsic work values since they both place great importance on the same dimensions. Studies have provided evidence that when a discrepancy exists between the values of the leader and the values of the follower it has negative follower outcomes, such as absenteeism and a decrease in job satisfaction (Mitra et al., 1992). As transformational leaders are so focused on intrinsic rewards, they will not respond to the extrinsic work values of highly Millennial identifiers and therefore the levels of extrinsic work values will decrease. So when transformational leadership is perceived as high by employees, the relationship between Millennial identification and extrinsic will become weaker, resulting in lower levels of extrinsic work values. In turn, as transformational leaders influence the extrinsic work value levels negatively, this will lead to a weakening of the expected negative relationship between extrinsic work values and job
satisfaction, in which job satisfaction levels will increase.
H9: Transformational leadership moderates the mediated relationship between Millennial identification and job satisfaction through extrinsic work values, such that when
transformational leadership is high (vs. low) Millennial identification will have a less positive effect on extrinsic work values, which in turn will positively relate to job satisfaction.
2.3.2 Transactional leadership
Transactional leaders participate in a basic exchange process in which they specify the expectations to the followers and reward them for their performance. So there is no deeper understanding of their values or development of their needs, but transactional leaders work “with followers' current needs and try to satisfy those needs with desired outcomes once agreed upon performance levels are achieved” (Jung & Avolio, 2000, p. 951). Referring back to the values of Millennials, expecting that they have higher levels of extrinsic work values, they will place great importance on being rewarded extrinsically for their performance. Exactly in line with the exchange leadership style of a transactional leader. As a transactional leader responds to the extrinsic work values of Millennials this will be increased. Therefore transactional leadership will influence the positive relationship between Millennial
identification and extrinsic work values, such that especially when transactional leadership is high the relationship between Millennial identification and extrinsic work values will be strong. In turn, as transactional leaders positively influence the extrinsic work value levels, this will lead to a strengthening of the expected negative relationship between extrinsic work values and job satisfaction, in which job satisfaction levels will even further decrease.
H10: Transactional leadership moderates the mediated relationship between Millennial identification and job satisfaction through extrinsic work values, such that when
transactional leadership is high (vs. low) the effect of Millennial identification on extrinsic work values will be especially strong, which in turn will negatively relate to job satisfaction.
On the contrary, since transactional leaders are so focused on extrinsic rewarding
they will not positively affect those with intrinsic work values, as they desire for example intellectual stimulation instead of status increase. This will not be achieved by this basic exchange process. As with transformational leadership and extrinsic work values, a
discrepancy between leader and follower’s needs will occur when Millennials with intrinsic work values have a transactional leader. So transactional leadership influences the negative relationship between Millennial identification and intrinsic work values in such a way that when employees perceive high transactional leadership, the negative relationship between Millennial identification and intrinsic work values will be weakened. Consequently, as transactional leaders influence the intrinsic work value levels negatively, this will lead to a weakening of the relationship between intrinsic work values and job satisfaction, in which job satisfaction levels will decrease.
H11: Transactional leadership moderates the mediated relationship between Millennial identification and job satisfaction through intrinsic work values, such that when transactional leadership is high (vs. low) the effect of Millennial identification on intrinsic work values will be especially weak, which in turn will negatively relate to job satisfaction.
3. Method
This chapter describes the empirical section of this study. First the research model is visualized, second the characteristics of the collected sample are described and lastly variables included in the questionnaire with corresponding reliability measurements are outlined. The complete questionnaire (in Dutch) is attached in the appendix.
3.1 Research design
The aim of this paper is to study how intrinsic and extrinsic work values of Millennials influence job satisfaction. This is achieved by an explanatory study in which the relationship between Millennial identification and job satisfaction can be explained through mediation of intrinsic and extrinsic work values and by the influence of transformational and transactional leadership. In order to answer the proposed research question the actual design is visualized as followed (see Figure 1):
How do intrinsic and extrinsic work values mediate the relationship between Millennial identification and job satisfaction and to what extent do transformational and transactional leadership moderate this relationship?
Figure 1. Research Model
In order to explain these related concepts, answer the research question and test the hypothesis, this study made use of a cross-sectional quantitative survey in which the data was collected from different groups of people; men and women of different ages and with different educational backgrounds (Saunders & Lewis, 2012). A cross-sectional study makes it possible to perform a study in a short amount of time while obtaining a reasonably large amount of data (N=149). Also, an online survey makes it easy and quick to administrate, limits observer bias and by letting participants fill them in anonymously it will also limit subject bias, which increases reliability (Saunders & Lewis, 2012). Additionally with this design, survey testing issues are limited as the data collection process is not subjected, leading to increased validity (Saunders & Lewis, 2012).
Nevertheless, this approach also has its weaknesses. For one such a study also
described as a correlational study doesn’t prove causation, meaning that the relationships of the variables only prove correlation. Second, a quantitative study also has internal validity
weaknesses. With an experimental study manipulation is performed in which you eliminate other factors that might affect the results of the study. As this study is not an experimental design it is possible that the results have been affected by other factors not included in this study, which in turn weakens the internal validity. Nevertheless a survey design was chosen as it allows for a strong external validity.
Indeed, it facilitates a larger sample size resulting in a higher generalizability of the
population, assigning more power to the conclusions and the proposed recommendations. In line with this, by using surveys the ecological validity increases compared to an
experiment, because the situation is more in line with daily life settings. For example a respondent can take the survey in their own situation whenever they choose to, instead of being restrained to an artificial context as is the case with an experiment. In order to strengthen this effect the survey was distributed in Dutch -the native language of the employees- and the introduction was written by an employee to conform to the culture of the company. Finally, a survey allowed the collection of a greater number of participants (N=149) within a short time frame (3 weeks), which is essential in explaining the relations between all the variables (Saunders & Lewis, 2012).
Surveys were distributed per email and could be filled in online from 23th of March
until 13th of April 2018. Participation was voluntary and the introduction of the survey contained an explanation of the purpose of the study, the procedure and how the obtained data will be handled. Also total anonymity was guaranteed to decrease socially desired answers. After the introduction a few demographic questions were asked to let the
participant become familiar with the survey, followed by the questions of the key elements (Millennial identification, intrinsic and extrinsic work values, transformational and
questions. The duration of the questionnaire was approximately 10 minutes. Before sending out the questionnaire to 230 employees a pilot was created with three employees and the questionnaire was altered according to their feedback.
3.2 Sample
The population consists of employees who identify themselves as Millennials, which was validated by questions in the survey. In order to explain how work values mediate the relationship between Millennials and job satisfaction and also how this is moderated by leadership styles, only followers were studied. So the population consists of all followers in the Netherlands who identify themselves as Millennials. In order to restrain the study from bias, a simple random sampling technique as part of probability sampling is used, which will increase validity and generalizability (Saunders & Lewis, 2012). The study has been held in an international recruitment agency, who delivered a list with 230 followers, who were invited by email to participate in the online survey. The estimated respondent level of this sample of 230 followers will be around 50% leading to 115 respondents. According to Green (1991) this meets the minimum requirements for the sample size, as the rule of thumb consists of 50+8K=58. In this study k=1 (independent variable) which implies that the sample should consist of at least 58 respondents.
In total 149 respondents filled out the questionnaire online which made the response
rate 64.8%, higher than expected. Of the 149 respondents (Mage = 27.2, SDage = 6.0, age range: 19-47 years) the majority (75.8%) were female. A large part of the sample (55%) has a higher professional education (HBO), 6.7% has a University degree, 17.4% a secondary vocational degree (MBO) and 20.8% has only a secondary educational degree
(VMBO/MAVO=1.3%, HAVO=16.8%, VWO=2.0%) and a remaining 0.7% had no educational degree at all.
The majority of the respondents work fulltime (67.1%) and only 8.7% work 24 hours
or less per week. A total of 20.1% of the respondents reported an organizational tenure of less than one year, 59.8% of the respondents reported an organizational tenure of 1-5 years, 12.8% has an organizational tenure of 6-10 years, 4% reported an organizational tenure of 11-15 years and 3.4% reported an organizational tenure of 16-18 years. The majority of the sample (58.4%) has a job tenure of 5 years or less.
3.3. Measurements
The surveys were constructed in Qualtrics and sent out digitally to followers. All the
questions containing the continuous variables could be answered by followers on a 5 point Likert Scale, which produced ordinal data to analyze. A Likert Scale is chosen as this has the advantage that followers are not forced into a simple yes or no answer, but allow them to have degrees of attitudes or opinions. However, the disadvantage of this scale is that followers are more likely to give socially desirable answers, in order to prevent that total anonymity was guaranteed.
3.3.1 Translation
All items included in the survey were derived from studies published in English journals. Since all the respondents are native Dutch speakers the original items were translated into Dutch. In order to assure that the content of the items remained unchanged, a parallel translation was performed, meaning that two independent native speakers translated the questionnaire into Dutch (Harkness, Vijver, & Mohler, 2003). Parallel translation will help to
reveal idiosyncratic wording, different interpretations and clear cut errors that occur in translations (Wolf, Joye, Smith, & Fu, 2016). Both translations were compared and based on conciseness, fluency, meaning and questionnaire design the most suitable questions were chosen, generating a new questionnaire in Dutch (Wolf et al., 2016). This method was chosen over the commonly used back translation since there is some critique about the quality of this translation method, as it is only done solely and because it supports a
translation that emphasizes equivalence of design over equivalence of meaning (Harkness et al., 2003; Wolf et al., 2016). One variable (transactional leadership) was already developed and validated in Dutch by de Hoogh, den Hartog, & Koopman (2004) and therefore excluded from the parallel translation.
3.3.2 Scale reliability
In order to test the reliability of the scale Cronbach’s Alpha was computed for each variable. All scales were reliable (α > .70, see below). The corrected item-total correlations indicated that all items have a good correlation with the total score of the scale as they were all above .30. Also, none of the items would substantially affect reliability (> .10) when deleted so none of the items was deleted.
Millennial Identification – independent variable
Based on their six item scale Mael & Ashforth (1992) test the perception of an individual’s belongingness to an organization, known as organizational identification. Since Millennial identification (to my knowledge) has not been used before as measurement in surveys, organizational identification was used and modified by replacing the word “organization” with Millennials. Mael & Tetrick (1992) extended these six items to 10 items testing for organizational identification, which was used to test our Millennial identification. However,
as the 10th item had a low factor loading (.190) in the study of Mael & Tetrick (1992), it has been chosen to delete this item for this study resulting in nine items testing Millennial identification. Followers could answer to what extent they agreed with the statements, an example of such a question is “When someone criticizes Millennials, it feels like a personal insult”. One item was reverse coded “I don’t act like a typical Millennial”, meaning that when the score is relatively low on this item Millennial identification is considered relatively high. The Cronbach alpha of this scale is .78 which indicates that these nine items are reliable (Field, 2013).
Transformational leadership - moderator
Transformational leadership is measured using the scale of Rafferty & Griffin (2004). Their 15 items distinguish the five important factors of transformational leadership; vision,
inspirational communication, intellectual stimulation, supportive leadership and personal recognition. Followers could answer their agreeableness to the statements given about their leader such as “my leader says positive things about the team”. One item was reverse coded, meaning that a relatively low score on this item refers to a relatively high score of
transformational leadership, the item is “has no idea where the organization is going”. Considering the Cronbach alpha (α = .89) it can be concluded that the reliability is high.
Transactional leadership - moderator
Transactional leadership is measured using six items from the transactional leadership section of CLIO (charismatic leadership in organizations) by de Hoogh et al., (2004). This measurement is chosen over the most common “Multifactor Leadership Questionnaire” (Bass, 1985), as the MLQ doesn’t fully acknowledge the exchange relationship between
leader and followers which should form the core of “transactional” (de Hoogh et al., 2004), whereas CLIO does emphasize the aspect of the exchange relationship between leader and followers. The statements about the follower’s leader could be answered with
agreeableness, an example of such a statement is “Attaches great importance to clear agreements and fair pay”. The reliability of these six items is considered very good (Cronbach’s α = .84).
Intrinsic and extrinsic work values - mediators
The intrinsic and extrinsic work value questions were introduced with the following;
“Different people may look for different things in their work. Below is a list of some of these things. Please read each one, then indicate how important this thing is for you.” This
measurement consisted of 10 items, six for intrinsic work values such as “A job which is interesting to do” and four for extrinsic work values such as “A job that has high status and prestige”. Originally Marini et al., (1996) included one additional item for intrinsic work values “a job where you do not have to pretend to be a type of personal that you are not”, however it had a weak loading on the intrinsic value construct (.430) (Marini et al., 1996) and therefore was excluded. Both variables were separately analyzed in this study, leading to a weak reliability for intrinsic work values (Cronbach’s α = .59). One of the corrected item-total correlations of intrinsic work values was below 0.30, which suggests to delete this item to increase the reliability, however the total scale reliability wouldn’t increase substantially when deleting this item (an increase below 0.10). Therefore it was chosen to retain the item. It is important to mention that this overall somewhat weak internal reliability of intrinsic work values could have consequences for the reliability of the results. Therefore the results of analyses containing intrinsic work values should be interpreted with caution. The internal
reliability of extrinsic work values was considered fine (Cronbach’s α = .74).
Job satisfaction - dependent variable
Job satisfaction is measured with eight items used from the scale of Bettencourt, Gwinner & Meuter (2001). Followers can indicate their satisfaction levels on statements such as “The amount of job security you have”. The Cronbach alpha of these items is .77, so it can be concluded that the reliability of these eight items is good.
Control variables
The control variables used are gender, age, tenure and job complexity. Gender is controlled for since a number of studies have shown that women have higher levels of job satisfaction compared to men (Blanchflower & Oswald, 1999; Brown & McIntosh, 2003; Clark, 1997). This is quite paradoxical since the labor market conditions are in general less favorable for women than for men (Sousa-Poza & Sousa-Poza, 2000). On the contrary there are also studies which show that men are more satisfied than women, either way it proves the importance of controlling for gender, which has been done in this study (Brush, Moch, & Pooyan, 1987).
Brush et al., (1987) argued with their meta-analysis of 19 studies that both age and
tenure are correlated with job satisfaction and therefore recommend to control for both. Kacmar & Ferris (1989) even found a curvilinear relationship between age and job
satisfaction. Due to the relationship between age and job satisfaction and organizational tenure and job satisfaction this research controlled for age and organizational tenure. All three variables are measured with single questions “What is your age?”, “What is your gender?” and “How long do you work for this organization?”.
Stogdill's (1974) leadership study showed that complexity of the job could affect employees job satisfaction. This effect exists because a job that has complex tasks requires numerous skills and is therefore mentally more demanding and challenging, which will most likely lead to more satisfaction (Morgeson & Humphrey, 2006). Job complexity is measured with four items from the Work Design Questionnaire by Morgeson & Humphrey (2006). These items were considered highly reliable (α= .80). All the items are reverse coded, an example of such a statement is “The job involves performing relatively simple tasks” in which agreeableness could be indicated.
4. Results
This section starts by describing the steps taken before testing the hypothesizes, followed by an overview of the correlations. Second the regression, mediation and moderation analysis will be reviewed and lastly the total moderated mediation will be described.
4.1 Preliminary steps
In total 149 respondents were collected, so the desired sample size of 58 was amply
achieved and therefore the requirements were met (Green, 1991). First a frequency test was run for all the variables to check for errors, there were no missing values. Second the data was checked for counter indicative items, which were present. The variable job complexity had four reversed items, Millennial identification one and transformational leadership contained one reversed item, all were recoded.
To check for normal distribution, kurtosis and skewness were calculated. Of the
seven variables only Millennial identification was normally distributed. Relatively high scores were obtained on the variables of intrinsic and extrinsic work values, transformational and transactional leadership and job satisfaction. Also the scores on the control variable job complexity were relatively high. The absence of a normal distribution amongst all these variables could be due to the fact that all the participants belong to one organization, all executing similar high educational level jobs. For instance, high job satisfaction levels can be explained as the organization is operating in a highly competitive market. This means that it is easy for an employee to switch to a new job when they are not satisfied anymore, so employees who are not satisfied do not stay at the company and are therefore not included in the sample. This all leads to an overestimation of the effects found in this study.
After that, standardized scores were created to identify outliers, in which an outlier was considered such if the standardized score was above 3.29 following the guidelines of Tabachnick & Fidell (2006). Instead of removing all these identified outliers, first a logical explanation was sought and after careful consideration the outlier was removed if such an explanation couldn’t be found. In total 11 outliers were removed.
In order to test the reliability of the scale Cronbach’s Alpha was computed for each
variable, as mentioned in chapter 3.3. All scales except for intrinsic work values (α = .59) were reliable (α > .70).
After checking the reliability of the scale, means were computed for Millennial
identification, intrinsic work values, extrinsic work values, transformational leadership, transactional leadership, job satisfaction and job complexity. With those means and the control variables gender, tenure and age a correlation matrix was created. The means, standard deviations and correlations of all these variables are shown in table 1.
As can been seen in the correlation matrix all the main variables; Millennial
identification, intrinsic work values, extrinsic work values, transformational leadership, transactional leadership and job satisfaction have significant and positive correlations. The highest correlation is those between transformational and transactional leadership
(r = .73, p <.01). More interesting is that both transformational leadership (r = .56, p <.01) and transactional leadership (r = .52, p <.01) have a significant and positive correlation with job satisfaction. Intrinsic and extrinsic work values have a significant positive relationship (r = .45, p <.01). Followed by Millennial identification which has a significant and positive correlation with intrinsic work values (r = .28, p <.01) as with extrinsic work values (r = .22,
p <.01), however both intrinsic and extrinsic work values (mediators) don’t have a significant
significant positive correlation, between intrinsic work values and transformational leadership (r = .17, p <.05).
Looking at the control variables, gender has no significant correlation with any variable, followed by tenure which has only one small correlation with a main variable;
transformational leadership (r = -.23, p <.01). Therefore, those two control variables will be excluded from further analyses. Since age and job complexity do have multiple significant correlations with main variables both will be included in further analyses.
4.2 Direct effects
A hierarchical multiple regression analysis was performed in order to examine the
relationship between Millennial identification and job satisfaction, when controlling for age and job complexity.
In the first step of the regression analysis the two control variables (age and job complexity) were entered. This model was statistically significant, F(2,146) = 13.79, p <.001, and
explained 15.9% of the variance in job satisfaction, meaning that the control variables age and job complexity do have a significant relationship with job satisfaction. Thereafter the independent variable Millennial identification was entered in step two. The total variance explained by the model as a whole was 16%, F(3,145) = 9.19, p <.001. Only an 0.01% additional variance in job satisfaction was explained, when adding Millennial identification to the model and controlling for age and job complexity (R2 change = .01, F(1,145) = 0.17,
p = .683). So Millennial identification has a non-significant relationship with job satisfaction.
In the total model only job complexity as a control variable was statistically significant (p <.001) with a Beta of .41, meaning that if a person’s job complexity increases with one their job satisfaction will increase with .41. Since the direct relation is not significant hypothesis 1 is rejected. An overview is given in table 2.
4.3 Mediation effects
The first set of hypothesis (2, 3 and 4) proposed that extrinsic work values would mediate the relationship between Millennial identification and job satisfaction. Hypothesis 5, 6 and 7 build upon the proposition that job satisfaction of highly Millennial identifiers is mediated through intrinsic work values. In order to test the two mediating roles model 4 of the PROCESS SPSS macro of Hayes (2017) was used. As mentioned before, the sample could have some normality assumption issues, therefore bootstrapping was used as this is a nonparametric resampling procedure for testing mediation without assuming a normal sample distribution (Preacher & Hayes, 2008). As recommended by Preacher & Hayes (2008) it was resampled 5,000 times in order to give an empirical approximation of the sampling distribution, which then is used to establish the confidence intervals of the indirect effect. First the mediating role of extrinsic work values between Millennial identification and job satisfaction was tested when controlling for age and job complexity. The results can be found in table 3 and table 4.
