The Influence of Age-Stereotyping by the Leader:

(1)

The Influence of Age-Stereotyping by the Leader:

The Impact of Age Congruence on the Exchange Relationship and the Job Performance Rating

Name: Floris Lucas Peter van der Weide Student number: 13084194

Date: June 26, 2022 Version: Final

EBEC approval number: EC 20220219080240

Executive Programme in Management Studies – Leadership & Management Track University of Amsterdam / Amsterdam Business School

Supervisor: Emma Op den Kamp

(2)

Statement of Originality

This document is written by Student Floris van der Weide (13084194) who declares to take full responsibility for the contents of this document. 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.

Signature ___________________________________________

(3)

Abstract

The current research is about the relationship between leader age-stereotyping and the job-performance rating of the follower. Moreover, I did investigate the influence of the exchange relationship between the leader and the follower and the possible impact of the degree of age congruence between them. I expected that when a leader has a high degree of age-based stereotypical perceptions, this would be negatively related to the job performance rating of the follower via a decrease in their exchange relationship, especially under a low degree of age congruence. In contrast, I hypothesized that this negative relationship is tempered - or may even turn positive - under a high degree of age congruence. I expected the degree of age-congruence to be of influence due to the leader’s process of age-based in-group and out-group formation of his followers. In this thesis, I could not find empirical evidence among 71 dyadic couples of leader and follower that age-stereotyping by the leader negatively impacts the job performance rating. The expected moderating role of age congruence did not have the presumed influence on the job performance rating and exchange relationship. Moreover, I could not find statistical evidence for the leader-member exchange mediating role. On the contrary, the survey results showed significant correlations between leader age-stereotyping and the job performance rating and the self-report of the follower on the job performance rating. Additional analyses supported the influence of leader-follower age congruence on the relationship between leader age-stereotyping and the follower’s self- evaluation of the job performance. The proposed suggestions for future research could expand the perspective on the impact of age-stereotyping by the leader. I recommend having a closer look at the impact of age-stereotyping on the self-image of the follower and the impact of age congruence between leader and follower within organizations.

(6)

1. Introduction

Pigeons tend to be common birds, but they represent several symbolic meanings. For example, the white pigeon symbolizes peace and love. Most ancient civilizations, such as the Egyptians, Romans, and Greeks, honored pigeons and identified them with life, fortune, and luck (Alois, 2020). ‘Pigeonholing’, on the other hand, represents a far less positive meaning, symbolizing the process of the categorization of groups, primarily based on unfair stereotypes. Stereotypes are “the beliefs about the characteristics, attributes, and behaviors of members of certain groups” (Hilton & Von Hippel, 1996, p. 240), and one can form stereotypical beliefs based on various factors, such as, e.g., race, religion, gender, or age.

The Dutch population is aging and has to work longer to gain enough money for the retirement funds and the national health care system (Serkozy, 2020). Moreover, the average age increases, as does the percentage of elderly employees in the workforce (HR-kiosk, 2017). Therefore, more generations than before are simultaneously at work (Lester et al., 2012). The aging of the Dutch population implies an increasing age diversity due to the broader age range within organizations (De Meulenaere et al., 2022). Consequently, there are increasing age differences on the work floor. Most leaders (35%) are 35-45 years old (ISBW, 2020). Most male followers are between 50 and 54 years old, and the age category of 55-65 is the fastest-growing category for women (HR-kiosk, 2017). This implies that, more often, leaders and followers differ in age. I define the degree to which people differ in age or are similar in age as ‘age congruence’. More specifically, the leader-follower age congruence (LFAC) is the degree to which a leader is close in age to his follower. As there are currently more differences in age on the workfloor, this implies that more often, the LFAC is at a low degree.

(7)

The aging of the workforce increases the risk of age-stereotyping in organizations (Čančer et al., 2016). Especially when LFAC is low, leaders form stereotypes based on age, as leaders and followers are less related to each other in terms of age. Moreover, this may influence the followers' well-being and performance. Age-based stereotypes can result in biased treatment of individuals (Finkelstein et al., 2015). For example, research by Lamont et al. (2021) suggests that biased treatment by leaders threatens the performance and well-being of older followers. Moreover, the impact of leader age-stereotyping (LAS) on the job performance process is influenced by the exchange relationship quality between leader and follower. I propose that when there is a low degree of LFAC, LAS will result in lower exchange relationships and, therefore, lower performance. On the contrary, when there is a high degree of LFAC, LAS is less or even not negative in its impact on the exchange relationship.

The leader-member exchange (LMX) model is one of the essential leadership models for exchange relationships (Gooty & Yammarino, 2016). Leaders establish stronger exchange relationships with a part of their followers (the "in-group") compared to other followers (the

"out-group") (Dansereau et al., 1975). Age is an essential demographic characteristic in forming the in-group and out-group, affecting in-group favoritism as followers close in age to the leader - i.e., high degree of LFAC - are more likely to be part of their in-group (McNamara et al., 2016). Studies have shown the effect of age in forming a leader's in-group and out-group (Hortense & Van Rossem, 2021; Van Rossem, 2019). In-group members are more favored than out-group members. This 'in-group effect' stimulates the quality of their interactions, which enhances LMX. Leaders may even provide more favorable job performance ratings (JPR) to their in-group followers (Clarke et al., 2019). I expect high LAS to be negatively related to the JPR of the follower via a decrease in LMX, especially under

(8)

low LFAC. In contrast, I hypothesize that this negative relationship is tempered - or may even turn positive - under high LFAC, that is, when the leader and follower have a high degree of age congruence.

Summing up, I aim to examine how leader age-stereotyping (LAS) influences the job performance rating (JPR). Moreover, I investigate the potential mediating role of the exchange relationship between leader and follower (LMX) and the potential moderating impact of leader-follower age congruence (LFAC) on the relationship. See Figure 1 for a visual overview of the conceptual model of this thesis.

The thesis is based on quantitative research and examines the above main variables in leader and follower dyadic couples. The research is performed by using an online survey sent to the professional network of the author of this thesis. The gathered sample for this study consists of professionals from various industries and companies within the Dutch workforce.

The relevance of this research is explained by a thorough review of the existing literature and followed up by a description of the method used to perform this research. Finally, the results are statistically explained and discussed in detail with possible limitations, managerial implications, and suggestions for future research.

Figure 1

Conceptual model

(9)

2. Literature review

2.1 Leader age-stereotyping

2.1.1 Heuristics

In 1905, Albert Einstein published his first and even Nobel prize-winning paper about heuristics. Heuristics comes from an ancient Greek word meaning ‘to discover’ and is an approach for solving problems when information is incomplete or even false. The concept developed over the years, and in the 1950s and 1960s, the meaning of heuristics became known as shortcuts; rules of thumb to guide the search for solutions in challenging situations (Hertwig & Todd, 2002). Heuristics may be pretty helpful in some situations as they may facilitate how one forms a judgment about certain events. Therefore, such ‘judgemental heuristics’ are beneficial for guiding decision-making in complex environments. Stereotypes are a form of judgemental heuristics (Souchon et al., 2009). Indeed, stereotypes may similarly steer someone’s judgment in certain situations.

2.1.2 Age stereotypes

Stereotypes are the responses to others based on general beliefs about the members of certain groups and on the characteristics and behaviors of these group members (Bodenhausen & Wyer, 1985; Hilton & Von Hippel, 1996). One can form stereotypical beliefs about groups based on various attributes, such as culture, race, religion, gender, generation, and age. The aging of the workforce leads to an increased risk of separating into subgroups based on age (Harrison & Klein, 2007). Individuals notice the age of other persons as one of the first demographic characteristics, which substantially impacts their perceptions of other people (Kite et al., 1991). The disadvantage of stereotypes is that they are selective

(10)

and overlook individual differences, as these judgments are not valid (Souchon et al., 2009;

Stanton, 2017). Moreover, the aging of the workforce may promote stereotypes that may harm organizations (Čančer et al., 2016; Posthuma & Campion, 2009).

2.1.3 Age-stereotyping in organizations

Harmful effects of age stereotypes in organizations apply to younger and older employees, as these age-based stereotypes are vulnerable (Čančer et al., 2016; Finkelstein et al., 2015; Posthuma & Campion, 2009). Age stereotypes negatively affect work performance (Manzi et al., 2019). The individual could perceive these judgments as a lack of compatibility, due to his age, with the desired job profile. This may substantially reduce work performance and commitment, and participation in the organization's long-term goals. The aging of the Dutch population implies an increasing age diversity due to the broader age range within organizations, resulting in more age differences on the work floor and separation based on age (De Meulenaere et al., 2022). The age separation results in employees categorizing themselves and their colleagues into in- and out-groups (De Meulenaere et al., 2016; Harrison & Klein, 2007). Employees identify themselves with others based on age, becoming part of their in-group (McNamara et al., 2016). Regarding age congruence, the degree of age congruence is high in the in-group but low in the out-group. In- and out-group members have negative stereotypes against the other group (Kunze et al., 2011).

Similarly, leaders tend to form their in-group of followers of comparable age, where LFAC is high. As a result, they hold negative stereotypes against the followers in the out- group with a low degree of LFAC. These age-based stereotypes make employees feel undervalued, excluded, and rejected due to being mistreated just because of their age (Kunze et al., 2011). With the aging population and the implication that there is an increase in low-

(11)

degree LFAC, increasing the leader’s out-group members, LAS could be a rising problem in organizations.

2.2 Job performance rating

2.2.1 Job performance rating

The degree to which employees perform well in their jobs is vital for the success of any organization. Job performance has been defined as “the total expected value to the organization of the discrete behavioral episodes that an individual carries out over a standard period” (Motowildlo, 2003, p. 4). Job performance consists of task performance and organizational citizenship behavior (OCB). Task performance is solely about performing the tasks of the daily job. OCB is about performing non-daily duties (Settoon & Bennett, 1996), such as helping colleagues or caring about others in the organization. Choudhary et al. (2017) argue that several factors influence the job performance in general, such as the nature of the job itself, the organizational setting, the amount of work one has to deal with, the reward system, and the perception of fairness. Moreover, the role of the leader is essential in affecting followers’ job performance.

In most organizations, the leader evaluates the follower's job performance. During the job evaluations, the leader analyzes and rates the follower's performance, the job performance rating (JPR). The JPR by the leader consists of evaluating the task performance of the follower and their OCB. These elements of the follower's performance and behavior influence the leader's evaluation (Byoung Kwon & Hyoung Koo, 2017). The JPR by the leader is based on the follower's performance, but there tends to be a risk of biased perceptions due to the age-based, in-, and out-group effect.

(12)

2.2.2 Bias in the job performance rating

The role of the leader in the JPR is crucial (Choudhary et al., 2017). However, the leaders may be biased in evaluating the follower's job performance (Anderson et al., 2015;

Huffcut & Roth, 1998; Kraiger & Ford, 1990). The JPR of the follower is subjective to the perspective of the leader. Leaders could think specific items of task performance or OCB are, in term opinion, at least expected and therefore do not weigh this in the JPR of the follower (Byoung Kwon & Hyoung Koo, 2017). On the other hand, leaders could be biased because their perception of their followers is based on their age. Age as a demographic characteristic is at the core of social identification with others (McNamara et al., 2016). Social identification is part of how people, in general, relate to group memberships. It starts with 'social categorization', forming an in-group and out-group, followed by 'social identification', the in-group becomes part of the social identity and de-identification with the out-group.

Finally, the 'social comparison' takes place where the in-group is better perceived than the out-group (Hogg, 2001).

Similarly, social identification is applicable for leaders in forming their followers into an in- and out-group based on age. The lower the age congruence between leader and follower, the more a follower is part of the out-group of the leader. McNamara et al. (2016) discovered that leaders rate the follower's job performance of the same age the highest. As Gooty and Yammarino (2016) agree, leaders tend to rate followers from their in-group higher than those from their out-group. The leaders' perceptions of age-based differences between their in-group and out-group direct stereotyping of their followers (Van Rossem, 2019).

Stereotypes about in-group followers tend to be more favorable than stereotypes about out- group followers (Hilton & Von Hippel, 1996). Because of the growing age separation in the

(13)

group of the leader. Therefore, there tends to exist an increasingly negative impact of age- stereotyping by the leader on the job performance rating of the follower; the first hypothesis is:

Hypothesis 1: There is a negative relationship between LAS and JPR.

2.3 Leader-member exchange

2.3.1 Exchange relationship

The interaction between leader and follower is based on their exchange relationship.

The leader-member exchange (LMX) model is one of the essential leadership models for exchange relationships (Gooty & Yammarino, 2016). The LMX theory defines leadership as developing individual exchange relationships between the leader and their followers. At the core of the LMX theory, leaders establish better relationships - higher exchanges - with a part of their followers, the in-group, than with other followers, the out-group. The exchange may take the form of followers approving the status of the leader and being loyal. The leader, in return, rewards the follower (Dansereau et al., 1975). This relationship may range from a lower to a higher level. When there is low LMX, the relationship between the leader and the follower may be solely based on the employment contract and the exchange, thus purely monetary. On the contrary, when LMX is high, the relationship goes beyond the employment contract, and the exchange is also non-monetary. In a high-quality exchange relationship with their leader, these followers are rewarded with respect, affection, support, loyalty, and commitment, which are social exchanges (Martin et al., 2016).

LMX has crucial tradeoffs for the success of an organization because it could stimulate followers to go beyond their employment contract. In a high LMX relationship,

(14)

followers are more motivated to go beyond their employment contracts. Followers tend to do so due to the reciprocity principle. Reciprocity is the process of paying back what we have received from others (Settoon et al., 1996). Social exchange is about reciprocity and concerns the motivation to perform beyond what is required. The more a leader provides his followers with rewards, the more the followers are inclined to strive for and achieve more than what is needed in exchange for these rewards. This process increases the positive feelings about the other, and the leader gains more affection and liking towards the follower. The follower is, as a result, more motivated to reach the leader's demands (Martin et al., 2016). Based on social exchange and reciprocity, followers invest more in their jobs to give in exchange to their leaders.

2.3.2 Leader bias and LMX

Researchers found that leaders vary in the extent to which they are biased with stereotypes about their followers. Stereotypes by the leader are based on the characteristics of the followers (Anderson et al., 2015). Neurological evidence shows that information about in- group members is processed more deeply than out-group members (Van Bavel et al., 2008).

The social identity theory of leadership is based on this process. This theory is about how people relate to each other and how they relate to group memberships. Moreover, it is about how leaders relate to followers of their in- and out-group, influencing their stereotypes about these followers. Stereotypes about the follower may impact the leader's evaluation of the elements of the exchange relationship with this particular follower. When the leader does not engage in exchange, it results in low-quality LMX (Chen et al., 2010). The biased perceptions may influence how the leader may evaluate the followers. Consequently, the leader may withhold exchange with followers that are part of their age-based, out-group, and are at a

(15)

low-degree LFAC. Followers that perceive adverse effects due to LAS could experience a negative impact in their experience of reciprocity. Moreover, they drop their motivation which negatively affects the exchange relationship with their leader (Kunze et al., 2017). This results in the second hypothesis:

Hypothesis 2: There is a negative relationship between LAS and LMX.

2.3.3 In-group favoritism

For an organization, high LMX relationships are vital for the organization's success.

LMX is related to the two elements of the job performance of the follower: task performance and OCB. Both types of job performance may be considered forms of exchange from the follower's side (Settoon et al., 1996). In addition, Ilies et al. (2007) argue that the interaction between leader and follower most often stimulates OCB. LMX positively affects employee performance (Martin et al., 2016). More specifically, leaders tend to give followers in high- quality relationships higher JPR compared to those in low-quality ones (Gooty &

Yammarino, 2016). In-group favoritism is preferring those from the in-group. Leaders evaluate those followers with higher JPR.

Different perceptions of the in-group and out-group exist. Stereotypes about in-group members tend to be more favorable than stereotypes about out-group members (Hilton & Von Hippel, 1996). The higher the degree of LFAC, the more a follower is part of the leader's in- group. Consequently, based on age, the biased perception can adversely affect the exchange relationship between leader and follower. Moreover, the relationship quality affects the JPR by the leader, as the leader tends to act more favorably towards the in-group followers. As a result, the third hypothesis is:

Hypothesis 3: The LMX mediates the relationship between LAS and JPR.

(16)

2.4 Leader-follower age congruence

2.4.1 Age differences

Reviews of the literature (e.g., Lyons & Kuron, 2014) found differences between individuals of certain ages regarding their personality, values at work, work-life balance, style of leadership, and preferences. Moreover, the historical events and the stage of an individual's life cycle tend to influence the personality (Lyons et al., 2015). The demographic changes within the workforce in most organizations result in challenges for companies due to the aging of their employees and the increase in age diversity (Kunze et al., 2016). There are opposing views about the impact of age differences within teams.

A positive effect of age differences is that it may foster collaboration within a team as it may enhance the variety in knowledge, skills, and abilities within a team. Moreover, age diversity provides more diversity in information processing, problem-solving, creativity, and the accomplishment of tasks (Bantel & Jackson, 1989; De Dreu, 2006; Kearny et al., 2009;

Van Knippenberg et al., 2004). The adverse effects derive from the social categorization process where team members form subgroups: the in-group (in this case consisting of others of the same age) and the out-groups (those of a different age) (Hogg, 2001; Richard & Shelor, 2002). As described in the previous chapter, intergroup bias leads to in-group favoritism, and consequently, members of the out-group are more negatively experienced. In support of this, Webber and Donahue (2001) found that people tend to feel more comfortable with people who are close to them in age. Moreover, there is less interaction with the out-group members, which disrupts the exchange of information and cooperation with those team members (Brewer & Kramer, 1985; Van Knippenberg et al., 2004; Wegge et al., 2008). This disruption

(17)

of information exchange and a negative impact on the cooperation with 'out-group' team members are adverse effects of age diversity on the workfloor.

The adverse effects of age differences could lead to problems in the workplace. They can have a negative impact as they may reduce collaboration between followers from different generations (Lester et al., 2012). Finkelstein et al. (2013) argue that misperceptions based on age differences could lead to problems when cross-age colleagues interact at work.

For example, younger employees are less positive about older people's work performance (Kornadt & Rothermund, 2011). Leaders are role models in the position to handle age-biased (miss)perceptions (Kleisner & Jahn, 2020). Leaders play an essential role in managing teams with diversity in age. Therefore, diversity in age and the congruence in terms of age between the leader and follower play a crucial role within organizations. There seems to be a difference in the interaction between leader and follower when they have a low degree of LFAC than when there is a high degree of LFAC.

2.4.2 Age congruence between leader and follower

Previous research examined the effects and concluded the importance of comparable demographic characteristics between leaders and followers and their impact on performance ratings (O’Reilly et al., 1989). The interaction between a leader and follower may be different when there is an age difference compared to when they are closer in age. For example, when the leader is younger than the follower, the leader often lowery rates the follower on his expertise (Van der Heijden, 2016). LFAC incongruence increases when the age difference between the leader and follower becomes more considerable (Van der Heijden, 2016). As limited research to date has taken into account the age of both leaders and followers, the

(18)

knowledge about the influence of LFAC within organizations tends to be insufficient (Lyons

& Kuron, 2014).

Summing up, the leaders' perceptions direct stereotyping of the in-group (high degree of LFAC) and out-group (low degree of LFAC). The in-group is most positively stereotyped and the out-group more negatively (Hortense & Van Rossem, 2021). Consequently, followers with a high degree of LFAC could perhaps experience less or no negative impact of LAS, as they are part of the leader's in-group. Leaders are more comfortable when close in age and rate those followers who are close(r) in age higher (Gooty & Yammarino, 2016). Leaders tend to favor the followers with a high degree of LFAC and exhibit less positive behavior towards followers with a low degree of LFAC, which influences the LMX. Therefore, the effect of LAS on JPR is possibly higher when LFAC is low. Moreover, the influence of a low-degree LFAC is that LMX tends to strengthen the negative impact of LAS on JPR. This results in the fourth hypothesis:

Hypothesis 4: The relationship between LAS and JPR, mediated by LMX, is moderated by LFAC such that the negative relationship between LAS, LMX, and JPR is weaker when LFAC is high in comparison to when LFAC is low.

2.5 Research question

Based on the literature review of the main variables of this thesis: leader age- stereotyping (LAS), leader-member exchange (LMX), job performance rating (JPR), and leader-follower age congruence (LFAC), there is an urge in the organizational context of further examining the gaps in the literature. Therefore this thesis aims to examine the main research question:

(19)

How does leader age-stereotyping influence the job performance rating by the leader, and is there a mediating effect of the leader-member exchange? Moreover, to what extent is the relationship between leader age-stereotyping and leader-member exchange, and the employee job performance rating moderated by leader-follower age congruence?

3. Method

3.1 Sample and procedures

Data for this study were collected in The Netherlands. Out of the entire population, the workforce in The Netherlands, dyads of leaders and followers working for diverse organizations were approached. The dyads were recruited through the author's professional network by using the non-probability sampling method, convenience sampling. Research by dyadic measurement is in line with other types of leader-follower research (e.g., Clark &

Mahadi, 2016; Gooty & Yammarino, 2016; Linen et al., 1993). Therefore, the current study as well had a quantitative approach and relied on a dyadic survey for both the leader and the follower. The survey was designed in Qualtrics. Participation was voluntary, and the anonymity of the participants was protected as no traceable personal information was gathered. The approach of this study was approved by the Ethics Committee Economics and Business (EBEC) of the University of Amsterdam, Amsterdam Business School, reference number EC 20220219080240.

The data was collected with an online questionnaire. In total, 317 individuals were approached with the request to fill in the online questionnaire, as a dyad, with their leader or follower. For every person I have approached, I registered if it was a follower or a leader;

64% were approached as followers, and 36% were as leaders. The reason for approaching

(20)

more followers than leaders was to avoid too much selection bias by the leader in choosing their favorite follower to participate in this research which could influence the outcome of LMX and JPR. Out of these 317 requests, the survey for leaders was filled in 121 times, and the follower survey was filled in 114 times. Every dyad received a code that both leader and follower entered in the survey to match the answers of the members of each dyad. After deleting the unfinished responses, for example when participants stopped responding after completing only the demographic questions, 97 times leader surveys and 99 times follower surveys remained. A further examination of the data showed that 18 leader surveys lacked a corresponding follower survey, and 22 follower surveys lacked a corresponding leader survey. Those have been deleted. Moreover, 4 double responses on the leader or participant side have been removed. After deleting these responses, preparing the data for final analysis led to a sample of 71 dyads (142 respondents).

Of the resulting sample size, 60% of the leaders were male, and 50% of the followers were females. The average age of the leaders was 43.5 years old (SD = 6.7), and the followers were, on average, 36.82 years old (SD = 8.5). The most significant part (31.9%) of the participants worked in a company with a size above 10,000 employees. The participants worked in various organizations from different industries, i.e., 38% in financial services, 15%

in business services, 13% in general services, 7% in the governmental sector, and 7% in trading. The average amount of years of service for the leaders was 22.1 years (SD = 6.93) for leaders and 15.2 years (SD = 9.6) for the followers. The question about the dyadic tenure between the leader and the follower differed too much in the answers from both leader and follower. Therefore the mean was calculated to compute this variable. The dyadic tenure between leader and follower was, on average, 2.99 years (SD = 3.02). In general, most

(21)

leaders (36%) interacted with their followers at least multiple times a week, and 30% had daily interactions with them.

3.2 Measures

The original measures were in English. As the respondents had Dutch as their native language, the survey was in Dutch; therefore, the ‘translation-back translation’ method was used to ensure all questions were asked correctly. The English items were translated into Dutch and translated back into English. Someone else then checked the translation. The original questions and statements in English can be found in Appendix A. The translated version, in Dutch, can be found in Appendix B. A measure was used to collect the data for each of the main variables.

Leader age-stereotyping. The degree of LAS measured how a leader generalizes followers based on their age. This was performed by asking leaders about their beliefs regarding several age-based statements—typical age stereotypes about younger and older employees. As there is no extensive literature on measuring the degree to which a leader is stereotyping based on age, there was no available measure. Therefore, the current measure was based on the measurement of gender stereotyping by Coronel and Federmeier (2016). As well as on the Semantic Misattribution Procedure (SMP) with a Cronbach’s Alpha of .82 (Ye

& Gawronski, 2017). In these gender stereotype studies, participants were asked to respond to gender-based statements. The participants had to evaluate these statements and indicate how typically they matched with their views on women and men. In line with this approach, for the current research, examples of age-based stereotypes were gathered from several peer- reviewed research papers (Finkelstein et al., 2015; Lester et al., 2012; Lyons & Kuron, 2014;

(22)

participants were asked to reflect on these age-based statements and indicate to what extent they (dis)agree with them. Sample statements are: Older workers are rigid and less flexible, and Young employees are selfish and irresponsible. In the survey, the leaders reflected on 15 age-based statements with a 5-point Likert scale, ranging from 1 = totally disagree to 5

= totally agree. The Cronbach’s Alpha was 0.78.

Leader-member exchange. For LMX, the follower’s perception of the quality of the exchange relationship with their leader was measured by the 11-item LMX scale of Liden and Maslyn (1998). The original response scale is a 7-point one; I followed Greguras and Ford’s (2006) approach, i.e., a 5-point Likert scale for consistency in response scales throughout the survey. All responses were scaled from 1 = strongly disagree to 5 = strongly agree. Sample items are: My supervisor would defend me to others in the organization if I made an honest mistake, and I do work for my supervisor that goes beyond what is specified in my job description. The Cronbach’s Alpha was 0.84.

Job performance rating. The rating by the leader on the job performance of the follower was measured with 10 items developed by Williams and Anderson (1991). This measure is a combination of in-role behavioral and extra-role behavioral statements. The original measure consisted of 14 items. Items are most accurate when applying a threshold of 0.70 (Field, 2018). Afterward, 10 items remained of the original measure of Williams and Anderson (1991). The response scale is a 5-point Likert scale, from 1 = never true to 5

= always true. One statement - Fails to perform essential duties - from the original measure is a reversed statement; for consistency purposes, this item is re-reversed into Performs essential duties. Sample items are: [this follower…] Adequately completes assigned tasks and Fulfills responsibilities specified in the job description. The Cronbach’s Alpha was 0.84.

(23)

Additionally, the followers were asked to rate themselves on these job performance items. This data provided information for additional explorative analyses. The analyses of these self-reports are performed in the Results section of this thesis. The Cronbach’s Alpha was 0.82.

Leader-follower age congruence. The measurement of the age congruence between leader and follower was based on the age difference method of Tsui and O'Reilly (1989). I asked both leaders and followers about their age and subsequently, subtracted the leader's age from the age of the follower for each dyad. The closer to "0", the higher their LFAC score.

For example, one dyad consisted of a leader who is 40 years old and a follower who is 45 years old; their LFAC score is 5. For another dyad, the leader's age was 35, and the follower's age was 33; the age difference is -2. The LFAC is higher for the second couple as it is closest to 0.

4. Results

4.1 Correlation matrix

Means, standard deviations, and correlations of this thesis variables are presented in Table 1. As male leaders showed higher (M = 2.65) on LAS than female leaders (M = 2.32), the gender of the leader was used as a control variable. The gender of the follower -participants was precisely 50% and therefore also included as a control variable to simulate an equal distribution. Maslyn and Uhl-Bien (2001) found that tenure has been associated with the quality of LMX. Therefore the dyadic tenure is the last control variable that seems to be of interest for further analysis. The results on other possible control variables were not taken into account (e.g., industry and interaction frequency between a leader and follower) as there

(24)

was too much deviation in the results between the leader and follower within the dyadic couples. The results indicated that LAS was negatively correlated with JPR (r = -0.27, p <

0.05); the JPR value decreased as the value of LAS increased. LAS was also negatively correlated with JPR self-report (r = -0.27, p < 0.05); the value of JPR self-report decreases when the value of LAS increases. The gender of the leader was negatively associated with LAS (r = -0.39, p < 0.01); the mean score of LAS is higher for men (M = 2.65) than for women (M = 2.32); therefore, in this research, male leaders seemed to have had more stereotypical beliefs than female leaders. LMX was not significantly related to LAS, JPR, LFAC, and JPR self-report. The dyadic tenure was not correlated to any of the other variables.

Table 1

Means, Standard Deviations, Correlations

Note. N = 71 dyads (N = 70 dyads for JPR and N = 63 for Dyadic Tenure). Men are coded as 1, women are coded as 2.

**. Correlation is significant at the 0.01 level (2-tailed)

*. Correlation is significant at the 0.05 level (2-tailed)

(25)

4.2 Regression analysis

To test the hypotheses, a moderated mediation regression analysis was performed, Figure 2, Process model 8 (Hayes, 2013). The results of this analysis are presented in Table 2.

The control variables in this analysis were ‘gender of the leader,’ ‘gender of the follower,’ and the ‘dyadic tenure’.

Figure 2

Process Model 8

Note. To statistically test the relationships between the variables in a moderated mediation analysis, Process model 8 is used to run the analyses in SPSS.

Hypothesis 1 stated that there is a negative relationship between LAS and JPR. The results of the moderated mediation analysis showed that this direct relationship was not significant (c1’ = -0.05, p > 0.05). Therefore, there was no significant evidence for this hypothesis based on the statistics, which has been rejected. Hypothesis 2 proposed that there is a negative relationship between LAS and LMX. The results of the moderated mediation analysis showed that this relationship was insignificant (a1 = 0.2, p > 0.05). Therefore, there

(26)

was no statistical evidence of a relationship between LAS and LMX which resulted in the rejection of this hypothesis. Hypothesis 3 held that the LMX mediates the relationship between LAS and JPR. The effect of LMX on JPR is positive but non-significant (b1 = 0.08, p > 0.05). A mediation of LMX in the proposed relationship between LAS and JPR was not supported by statistical evidence as there were no significant results, and the hypothesis is therefore rejected. Hypothesis 4 suggested that the relationship between LAS and JPR, mediated by LMX, is moderated by LFAC. The negative relationship between LAS, LMX, and JPR is weaker when LFAC is high than when LFAC is low. The results indicate a negative but non-significant relationship (c3’ = -0.01, p > 0.05) between LAS and LFAC on JPR. Moreover, the moderation effect on the mediator (LMX) was negative and non- significant as well (a3 = -0.02, p > 0.05). After closer inspection of the conditional effects, both the direct and indirect effects were non-significant. The direct relationship between LAS and JPR is non-significant for LFAC. The degrees of LFAC were divided into 2.00, 7.00, and 15.92 years. The negative effects were marginal (-0.06, -0.09, -0.14). The indirect relationship of LAS and JPR via LMX was not contingent on LFAC (0.01, 0.00, and -0.01).

The lower and upper confidence intervals ranged from negative to positive for all the LFAC categories in the direct and indirect effect. In other words, the degree to which a leader uses age-based stereotypes does not significantly impact the rating of the job performance of the follower and is not affected by the degree of LFAC. Therefore, no significant statistical evidence was found that supported a moderating effect of LFAC on the relationships between LAS and JPR and between LAS and LMX. Concerning the proportion of the total variance of JPR, this model explains 33.14% and is not significant (p > 0.05).

(27)

Table 2

Moderated, mediation regression analysis

Note. Moderated mediation regression analysis results for leader age-stereotyping, leader-member exchange, leader-follower age congruence, and job performance rating.

4.3 Job performance rating self-report

Additionally, I asked the followers to self-report on their JPR. These results of the JPR self-report in the moderated mediation regression analysis are presented in Table 3. The overall conclusion is the same as the main conceptual model; there was no statistical evidence for a relationship between LAS and JPR self-report (c’1 = -0.16, p > 0.05).

However, the effect of LAS increased from -0.05 to -0.16; the relationship was also non- significant. Moreover, concerning the variance in the JPR self-report explained by LAS, this model explains 33% and is also not significant (p > 0.05).

(28)

Table 3

Moderated, mediation regression analysis

Note. Results of moderated mediation regression analysis for leader age-stereotyping, leader-member exchange, leader-follower age congruence, and job performance self-report.

4.4 Male leaders

The mean score on LAS for male leaders (M = 2.65) is higher than for female leaders (M = 2.32). Comparing the mean of both genders with an independent samples t-test provided significant results (p < 0.001). Therefore, I decided to examine these new insights further. I performed the same moderated mediation regression analysis to investigate this explorative research direction, only for the dyadic couples with a male leader. It resulted in a sample size of N = 40. Regarding the analysis with JPR as the outcome variable, the relationships remained non-significant. The impact of LAS on JPR increased to -0.20 (p > 0.05). Although the relationship is still non-significant, there was a clear improvement in the effect size.

Performing the same analysis, with male leaders only, with JPR self-report as the outcome variable, there was no improvement in the significance level of the relationship between LAS and JPR self-report. The effect size (-0.18) remained almost the same.

(29)

Additionally, some researchers (i.e., Wasserstein & Lazar, 2016) criticize the threshold of 0.05 as a significance level. According to them, a small sample with a higher p- value could still be important. Therefore, one could argue that despite the p-value (0.08), it seems to be that male leaders have more age-based stereotypical beliefs than female leaders, which could influence the follower’s JPR.

4.5 Adjusted leader age-stereotyping

The degree of LAS was measured by asking leaders about their beliefs regarding several age-based statements. This survey contained a total of 15 of these statements, of which 13 were negatively phrased, and 2 of the statements were phrased positively. After examining existing literature, Holleman et al. (2021) concluded that in decision-making, people have a different reaction to negative wording than to positive. Therefore, I decided to do some additional analyses concerning these new insights. I performed the same moderated mediation regression analysis to investigate this explorative research direction. To calculate the mean score for LAS, I only used the age-based statements, which were negatively phrased. Regarding the analysis with JPR as the outcome variable, the relationships remained non-significant (p > 0.05). I ran the same analysis with JPR self-report as the outcome variable provided interesting results. The non-significant relationship between LAS and JPR self-report remained non-significant (p > 0.05). Moreover, the conditional direct effect of LAS on JPR self-report changed into a significant effect (p < 0.05). Only the (negative) effect of a medium degree LFAC of 7 became significant (effect = -0.013, SE = 0.06 , CI:

-0.25 to -0.01). Therefore, one could argue that it seems to be that age-stereotyping by the leader significantly affects the JPR self-evaluation when the LFAC is moderate, with a degree

(30)

5. Discussion

This thesis aimed to examine the impact of LAS on JPR and discover the impact of LMX as a mediator in this relationship. Moreover, to what extent was LFAC of moderating influence. The results of this research show that there is no support for the proposed research question, which was:

How does leader age-stereotyping influence the job performance rating by the leader, and is there a mediating effect of the leader-member exchange? Moreover, to what extent is the relationship between leader age-stereotyping and leader-member exchange and the employee job performance rating moderated by leader-follower age congruence?

Regarding the non-significant results, there was no statistical evidence from this research to conclude the influence of the leader's age-based stereotyping on the follower's job performance rating. The results showed that the leader-member exchange seems not to have had a mediating effect on the relationship between leader age-stereotyping and the job performance rating. Moreover, in this study, there was no impact of age congruence between the leader and the follower on the relationships between leader age-stereotyping job performance rating, and the leader-member exchange. In addition, the explorative analyses provided some inspiration about the moderating role of age congruence between leader and follower. In what follows, I will outline some alternative explanations for the lack of statistical support, as well as some theoretical considerations and methodological limitations of the current study.

(31)

5.1 Limitations

5.1.1 Theoretical limitations

Based on earlier research, some theoretical limitations are possibly relevant to this thesis study. The sample and the measuring of JPR may have limited this research and, therefore, its results.

Interactions and tenure

Stereotypes may have their most considerable impact when the individuals forming the stereotypes have little contact with or do not have much first-hand knowledge about the target group (Murphy & DeNisi, 2021). When there is no or limited information about the target group, one tends to rely on general stereotypes about this target group. These stereotypes could be misleading or even false. The participants in this research had, on average intensive contact. 36% of the dyads interacted multiple times weekly, and 30% had daily contact. In other words, many of the samples had intensive contact and probably first- hand knowledge about the followers. In line with these findings, Van der Heijden (2016) found that stereotypes mostly apply when leaders and followers have a short dyadic tenure.

The dyadic tenure between leader and follower was, on average, three years, implying that overall the stereotypes did not have their most considerable impact.

Based on Murphy and DeNisi’s (2021) research about the frequency of contact and availability of first-hand knowledge, and in addition, to Van der Heijden's findings on the effect of short dyadic tenure on the influence of LAS, the current impact on JPR was less strong. In other words, if the leader has few interactions with the follower and relies on general stereotypes, which are often negative as the LFAC degree is usually low due to the

(32)

aging of the population, the impact on JPR is more substantial. As well as the influence on the mediating role of LMX, the exchange relationship between leader and follower would be less favorable due to limited interactions and not knowing each other well. The sample in the current research was, probably less suitable as the leaders and followers were high in their frequency of contact and had an average dyadic tenure. When the sample of dyadic couples had limited interactions and a shorter dyadic tenure, it probably would be more suitable for this type of research. It could have led to different results about the influence of LAS.

Biased perceptions

After reviewing the literature on performance ratings, Spence and Keeping (2010) found that there may be inaccuracies in the performance evaluations by the leader. So far, this thesis has brought to light that (age-based) stereotypes influence the JPR by the leader.

Spence and Keeping (2010) argue the main factors behind this are: 1) the possibility of adverse outcomes, 2) the norms within the organization, and 3) any possible self-interests of the leader. In the current research, I considered that leaders are biased in the job performance evaluation due to age-based stereotypes. As it may be, the results on JPR are more biased because of the factors Spence and Keeping found in their research. Therefore, relying on these possibly biased evaluations is maybe not an accurate indicator to measure the influence of LAS on JPR.

5.1.2 Methodological limitations

Several methodological limitations in the current study could have contributed to the reached outcomes. The main methodological limitations are the topics of the sample versus population and the leader-follower dyads.

(33)

Sample versus population

Due to the non-probability sampling method, it is unclear how close the sampling value approaches the population sample as convenience sampling does not provide evidence that they are representative of the population (Trochim et al., 2016). Moreover, LinkedIn was used to reach out to the participants. LinkedIn is currently used by around 5 million users in The Netherlands (Oosterveer, 2021), about 50% of the total workforce in The Netherlands. It may be possible that not every industry or demographic characteristic was represented similarly. It might be possible that the sample is not representative of the population. The industries in the survey were based on CBS (Dutch institute of statistical information) categories. Participants struggled with the categories that were too rigid and seemed to be outdated. Therefore, too much difference occurred in the answers within the dyadic couples.

This made it difficult to estimate and compare the sample with the population. The sample size of this study was not large, consisting of 71 dyadic couples. The sample size affects the significance as large samples have greater power to detect (significant) effects between variables (Field, 2018).

As a result, a limitation of this study is that there is no sufficient evidence that the sample is representative of the population. It may be possible that this had an impact on the results. Moreover, a larger sample size could also result in other and maybe significant statistical findings for the proposed model. Therefore, the results may only be generalized with great reservation.

Leader-follower dyads

Given that the responses were collected from leader and follower dyads, there is a risk of social desirability bias. Followers could fear that their responses could be viewed by their

(34)

leaders. Therefore, employees could be disposed to relay answers that they believe would be more favorable to their leader. Moreover, in terms of favorability, it is possible that leaders who agreed to participate in this study primarily chose their favorite followers to participate.

In support of this idea, the average job performance evaluation was relatively high (4.39 on a 5-point scale). In addition, followers may have often only asked their leaders to participate in this research when they were in a good relationship, similarly supported by relatively high LMX scores (4.11 on a 5-point scale). Furthermore, although the dyadic tenure between leader and follower seemed of interest in this research, it was a question in the survey with not too much accuracy in the answers by both the leader and follower. Possibly too many participants just estimated the dyadic tenure. Besides, common rater bias could occur as the same raters are used, at the same time, for all variables in the survey, which could affect the relationships between the variables (Podsakoff et al., 2012).

These dyad-related methodological limitations could have affected the outcomes of this research such that the results could be more reliable when the leaders and followers were in a less close relationship with each other. It may be possible that there is less impact of the in-group effect in that situation. Besides, if the dyadic tenure was more accurate, a better explanation of the results is possible. Likewise, the followers may be experienced the social desirability bias to a smaller extent.

5.2 Managerial implications

For organizations, there are some practical implications from this research. I propose several implications that perhaps will help organizations on the topic of age-stereotyping by their leaders.

(35)

Human Resources practices

The results suggest a significant correlation between LAS and JPR. Moreover, male leaders score higher on LAS than female leaders. Therefore, particularly male leaders, and leaders in general, need to be aware of the possible influence of biased perceptions about followers of certain ages on their leadership behavior. Gilbert and Hixon (1991) found that some processes can counteract the activation of stereotypes. The automatic responses can be successfully moderated by training to make counterstereotypical judgments (Coronel &

Federmeier, 2016). Age diversity should be actively managed by top management and Human Resources (HR) (De Meulenaere et al., 2022). Senior Management or HR could pay attention to this topic by offering training courses to provide leaders with tools to develop counterstereotypical beliefs, resulting in behavior to cope with these stereotypical beliefs.

Development of this behavior will ensure that leaders become aware of their stereotypical beliefs and can counteract this behavior, which may help mitigate adverse effects.

Additionally, HR practices, such as diversity and inclusion training, could foster the inclusion of age diversity in organizations. Moreover, it could reduce the age-based bias behavior of leaders (De Meulenaere et al., 2022). Therefore, organizations must guide their leaders in handling their beliefs and behavior.

The self-image of the follower

The results suggest that there may be an influence of LAS on the JPR self-report by the follower, which may indicate that the self-image of the follower is negatively influenced when the leader has a (high) degree of age-stereotyping. Moreover, after ruling out the positive phrased age-based statements, there seems to be a significant moderation effect of LFAC (moderate degree of 7 years) on the direct relationship between LAS and JPR self-

(36)

report. A negatively influenced self-image results in a lack of confidence in the job for the follower. The follower could get a negative view of the job-‘fit' as their age seems to be a problem for success in the job (Manzi et al., 2019). Enhancing the follower's view of fitting in with his job positively affects his attitudes, behavior, and performance (Edwards & Cable, 2009). Therefore, leaders must be aware of their behavior's influence on the follower's self- image. Moreover, they should help the follower feel compatible with his job instead of being negatively judged due to age-based assumptions.

Awareness of out-group characteristics

For leaders, it is essential to be aware of the characteristics, i.e., age, in forming their in-group and out-group. Moreover, it may be helpful when they notice the followers in their out-group. Not to exhibit negative behavior towards the out-group members but to actively involve these followers. Leaders aware of age differences and age-based stereotypes will reduce the conflicts on the work floor (Finkelstein et al., 2015). Diversity is a resource that increases followers' knowledge, skills, abilities, and opportunities and positively affects the interpersonal relationships within organizations (Thomas & Ely, 1996). Van Knippenburg et al. (2013) argue that a ‘diversity mindset’ is essential to look at diversity's positive effects and outcomes. When leaders become aware of the advantages of diversity on the work floor, a wide range of age diversity will become a valuable resource instead of a possible burden.

5.3 Suggestions for future research

Considering this study's limitations and managerial implications, there are enough reasons to expand research in this field of leadership. Therefore, I propose some suggestions

(37)

JPR self-evaluation

Concerning the additional results on the JPR self-report, there is a slight difference in the mean score by the leader (4.39 on a 5-points scale) and the self-report by the follower (4.32 on a 5-point scale). The self-report score in this research is lower than the leader rating, but there is insufficient statistical evidence (p > 0.05) for this explanation. Usually, followers evaluate themselves with more leniency than the leader (Heidemeier & Moser, 2009).

According to Manzi et al. (2019), age-based stereotypes negatively affect the individual as the follower could perceive the judgments based on age as a lack of fit with their job due to his age. It is also possible that the mechanism works the other way around. Kunze et al.

(2016) argue that followers who have a misfit with their job have a lower JPR and are possibly more sensitive in their perception of age-based discrimination. In this research, the LAS scores ranged from 1.40 to 3.47 (on a 5-point scale). When LAS was high, the negative impact on the self-image of the follower could be explained. It is unclear what the general cause for a higher JPR self-report score was. Moreover, to what degree the followers experienced a misfit with their job was not revealed in this research. Therefore, future research could enhance the view on this topic.

Significant correlations

Contrary to the conclusion of rejecting the hypotheses based on the non-significant results, the results indicated significant correlations between LAS, JPR, and JPR self-report.

These correlations may inspire future research to explore these constructs and their interrelationship further, perhaps using a different sample type and a larger sample size.

Probably this may as well be an influence on the results of the additional explorative research.

(38)

Descriptive and prescriptive stereotypes

This thesis focused on age-based stereotypes as the generalization of followers based on the description of their characteristics. It could be that the stereotypical beliefs leaders have are not only descriptive but also prescriptive. In that situation, it is also about the traits a follower of a certain age should have, according to the leader (Finkelstein et al., 2015).

Prescriptive stereotypes may play a more critical role in the evaluation by the leader than descriptive stereotypes, especially when a follower does not meet de prescriptive norms of the leader. The current study did not distinguish between descriptive and prescriptive stereotypes. Future research could examine the impact of the interplay of descriptive and prescriptive age-typical stereotypes on the evaluation of the follower.

Age versus generation

There tends to be much discussion within leadership literature about age-based thinking versus generational-based thinking. The disadvantage of generational stereotypes is that they are not always actual differences but are often the perception of individuals about possible differences between generations. These perceived differences lead to mistaken conclusions about generational differences. Therefore, it negatively affects collaboration between generations (Lester et al., 2012). Wong et al. (2008) argue that there are more differences related to age than generational differences and suggest not relying on generational differences but paying attention to other individual differences between followers. Therefore, age tends to be a better indicator than generations in exploring the effects of stereotyping by the leader on his followers. Rudolph et al. (2018) also propose not focusing on generational differences as they are exclusive, over-simplified, deterministic, and overemphasize the effect of cohorts. They suggest the focus on age. Therefore, for this

(39)

research, I chose to focus on age-based stereotypes instead generational-based stereotypes.

Generational stereotypes do not need to concur with age-based stereotypes, according to Van Rossem (2019). Other scholars argue that generations differ in several aspects (Benson &

Brown, 2011; Lyons & Kuron, 2014). Generalizing generational differences into stereotypes could even help leaders serve as a sense-making tool to push for deeper individual connections (Lyons et al., 2015). Therefore, future research focusing on generations may be helpful in the perspective of LAS with its impact on JPR and the mediating role of the exchange relationship between leader and follower.

Age-stereotyping by top management

LAS is especially harmful when it concerns top managers (Kunze et al., 2016). The explanation lies in the upper echelons theory; organizations are a reflection of their top-level management (Hambrick & Mason, 1984). The behavior of top management is perceived as the golden standard within an organization and is, therefore, extra harmful when it concerns LAS (Carmeli, 2008). Lower management in these organizations could perceive biased judgments based on age as appropriate behavior. Consequently, lower management may copy this behavior and exhibit these biased judgments toward their followers. Therefore, top management needs to know their role model position (Kunze et al., 2013). In the current research, there was no attention to the leaders' management level. In future research, more attention could be made to investigating LAS for the different management levels.

Age separation

Within organizations, age diversity can be organized in three ways; the separation of age (i.e., categorizing the workforce into homogenous, age-based groups), variety (i.e.,

(40)

ordering the workforce into heterogenous, age-based groups with plenty of variety), and disparity (i.e., categorizing the workforce into age-based groups in which one employee is much older than the others) (Harrison & Klein, 2007). It may be possible that the impact of LAS is different for each categorical setup of the workforce. For example, when the organization categorizes the workforce with disparity, the effect of LAS is perhaps higher than in a situation where the workforce is categorized into heterogeneous age-based groups.

In future research, the distinction could be made between dyads regarding the categorization of age diversity within their organization.

6. Conclusion

The current research aimed to provide statistical evidence for the relationship between leader age-stereotyping and job performance rating and the role of leader-member exchange and age congruence between leader and follower. Even though the current results did not support the proposed hypotheses, this research may inspire future thinking and research on this potentially valuable topic. Indeed, the significant correlations of LAS with JPR and JPR self-report create exciting opportunities to discover in future studies. The additional analyses may help other researchers to reveal the impact of age-stereotyping by the leader and the influence of age-congruence between leader and follower. The significant moderation effect of LFAC on the relationship between LAS and JPR self-report will hopefully motivate other scholars. Moreover, the methodological limitations of why other significant relationships were missing could be of inspiration for others. If age congruence plays a role in its impact on the interactions between leaders and followers in organizations, another study may reveal it.

(41)

References

Alois, R. (2020). Pigeon Symbolism: The Meaning And History Behind The Bird. Richard Alois. Retrieved from https://www.richardalois.com/symbolism/pigeon-symbolism Anderson, A. J., Ahmad, A.S., King, E. B., Lindsey, A. P., Feyre, R. P., Ragone, S., & Kim, S.

(2015). The effectiveness of three strategies to reduce the influence of bias in evaluations of female leaders. Journal of applied social psychology, 45(9), 522-539.

Bantel, K.A. & Jackson, S.E. (1989). Top management and innovations in banking: does the composition of the top team make a difference? Strategic Management Journal, 10(1), 107-124.

Benson, J., & Brown, M. (2011). Generations at work: are there differences and do they matter? The International Journal of Human Resource Management, 22(9), 1843-1865.

Bodenhausen, G. V., & Wyer, R. S. (1985). Effects of stereotypes on decision making and information-processing strategies. Journal of Personality and Social

Psychology, 48, 267-282.

Brewer, M. B., & Kramer, R. M. (1985). The psychology of intergroup attitudes and behavior. Annual review of psychology, 36(1), 219 - 243.

Byoung Kwon, C., & Hyoung Koo, M. (2017). Subordinates’ helping, voice, and supervisors’

evaluation of job performance: The moderating effect of supervisor-attributed motives. Career development international, 22(3), 222-240.

Čančer, V., Treven, S., & Rožman, M. (2016). Stereotypes of older employees compared to younger employees in Slovenian companies. University of Split, Faculty of Economics Management, 21(2), 165 - 179.

(42)

Carmeli, A. (2008). Top management team behavioral integration and the performance of service organizations. Group & Organization Management, 33, 712–35.

Chen, Z., Lam, W., & Zhong, J. A. (2010). Effects of perceptions on LMX and work performance: Effects of supervisors’ perception of subordinates’ emotional intelligence and subordinates’ perception of trust in the supervisor on LMX and, consequently, performance. Asia Pacific journal of management, 29(3), 597-616.

Choudhary, N., Naqshbandi, M., Philip, P. J., & Kumar, R. (2017). Employee job

performance: The interplay of leaders’ emotion management ability and employee perception of job characteristics. The Journal of management development, 36(8), 1087-1098.

Clarke, N., & Mahadi, N. (2016). Differences between follower and dyadic measures of LMX as mediators of emotional intelligence and employee performance, well-being, and turnover intention. European Journal of Work and Organizational Psychology, 26(3), 373-384.

Coronel, J. C., & Federmeier, K. D. (2016). The Effects of Gender Cues and Political Sophistication on Candidate Evaluation: A Comparison of Self-Report and Eye Movement Measures of Stereotyping. Communication Research, 43(7), 922-944.

Dansereau, F., Graen, G., & Haga, W. J. (1975). A vertical dyad linkage approach to leadership within formal organizations: A longitudinal investigation of the role- making process. Organizational Behavior and Hirinan Performance, 13, 46-78.

De Dreu, C.K.W. (2006). When too little or too much hurts: evidence for a curvilinear relationship between task conflict and innovation in teams. Journal of Management, 32(1), 83-107.

The Influence of Age-Stereotyping by the Leader: