Do differences in compensation schemes, social characteristics and gender explain differences in fairness perceptions?

(1)

Do Differences in Compensation Schemes, Social

Characteristics and Gender Explain Differences in

Fairness Perceptions?

Master’s Thesis

University of Amsterdam

14-06-2015

Thijs Verbij

5877547

MSc. BA

Supervision: Drs. A. el Haji

(2)

This document is written by Student Thijs Verbij 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.

(3)

Abstract

Employees would like to be treated fairly by their employers. When fair treatment is practiced in the workplace, employee effort and productivity can increase. For that reason, firms want to be perceived as being fair. Fairness perceptions are strongly dependent on compensation. It has been suggested that the compensation structure can influence people's opinions of the company. In addition, different social and individual characteristics and also different compensation schemes appear to explain the differences in perception. This study will research the differences in these fairness judgments by focussing on either individual and group payment methods, or variable and fixed payment methods. Confidence-level, social value orientation and gender also help to explain differences in fairness perceptions. Results confirm there are differences in fairness regarding compensation schemes. There is likely a connection between employee preferences and individual compensation when a variable payment structure is in place.

(4)

Index

1. Introduction ...5

2. Literature review ...7

2.1 Microfoundations/behaviour ...7

2.2 Overconfidence and social value orientation ...8

2.3 Fairness perceptions...11 2.4 Payment systems ...13 3. Hypotheses ...16 4. Methodology ...19 4.1 Sample...19 4.2 Method ...19 4.3 Research Design ...20 5. Results...22 5.1 Descriptive Statistics...22 5.2 Correlation...25 5.3 Regression ...26 5.4 Other Variables...28 6. Discussion ...31

6.1 Summary of the findings...31

6.2 Research implications ...32

6.3 Limitations ...32

6.4 Further Research...33

References...34

(5)

1. Introduction

Fairness perceptions are important because they are directly related to the effort that

employees put into their jobs (Stoelhorst & El Haji, 2012; Akerlof & Yellen, 1990; Cohn e.a., 2010). As a result, managers should optimize their employees' perceptions of fairness.

However, the heterogeneity of employees can make it difficult for managers to optimize these fairness perceptions. Employee compensation can also influence perceptions of fairness. Stoelhorst & ElHaji (2008) and Kahneman e.a. (1986) investigated the external factors that can influence employee compensation. In their study, factors such as firm profit or

unemployment and the actions that firms take regarding these factors, were tested and found to influence employee perception. In addition, the individual characteristics of the employee could be an important factor when this type of judgment is made. In this research, factors such as overconfidence, social value orientation and gender will be examined to determine whether they influence employees' perceptions of fairness. Oftentimes, an individual's estimation of his or her ability to perform their job exceeds their actual abilities. This important factor will be discussed later in this paper. Overconfidence is a possible indicator that there is a

mismatch between the results an employer expects from the employee and the performance that the employee delivers. This could indicate that the compensation the employee wants exceeds the actual compensation the employer will provide. This study will examine whether different compensation schemes influence the employee's perceptions of fairness, and if there are possible individual moderators that can affect these perceptions.

Overconfidence can be explained by looking at overplacement and overestimation (More & Healy, 2008). Overplacement occurs when an employee perceives him/herself to be better than another employee. Overestimation is occurs when an employee's perception of his/her own ability that exceeds his or her actual abilities. These two variables can influence whether or not the employee believes his compensation is fair.

Money as compensation implies a direct equal exchange between two parties (Zelizer, 1996). Generally, a certain degree of bargaining and accountability are exchanged between these parties.

(6)

As a result, it is difficult for a company to act in a correct, fair way with respect to the employee (Kahneman e.a. 1986; Burrows & Loomes 1994; Stoelhorst & ElHaji 2008). The research from Bridoux e.a. (2011) explains that employees have different motives for cooperating with each other, and different motives for putting more effort into the

organization. Focussing on micro-foundations, the individual characteristics of employees create a better understanding of why humans behave as they do. Furthermore, Felin & Foss (2005) indicated the importance of micro-foundations regarding the strategic management of an organization.

In addition to the overconfidence of employees, there are other individual

characteristics that can influence an employee’s decision making process (Van Lange, 1999). Nauta e.a. (2002) explained that social value orientation was linked with the likelihood that an employee had knowledge of the company and whether that employee was concerned about the goals of the company. To determine what factors influence an employee’s decision making process, this research will question whether the social value orientation has an effect on an employee’s fairness perceptions in combination with different compensation schemes.

Gender is the final variable that was tested for this study. This factor will be examined because research has shown that a relationship exists between males and females in combination with overconfidence (Barber & Odean, 2001).

The following questions will be answered in this research:

 Does overconfidence influence the decision about whether an employee’s compensation is considered fair or unfair?

 Do social value orientations influence an employee’s fairness perceptions when different compensation schemes are applied?

 Are there differences between male and female employees?

This paper will first discuss the current literature regarding microfoundations, fairness perceptions, overconfidence, social value orientation and payment systems. In chapter 3, the theoretical framework will be presented. This chapter also includes the hypotheses. The methodology and the research design will be explained in chapter 4. Chapter 5 will focus on the results of the study and in chapter 6, the findings will be summarized and discussed.

(7)

2. Literature review

2.1 Microfoundations/behaviour

Investigating whether overconfidence and social values are factors that explain behaviour begins with the microfoundations of the organization. Felin & Foss (2005, p. 1) mention that an organization would not exist without individuals. Approaching and analyzing individuals is key to understanding human behaviour, both in real-life and within a firm. The importance of using an individual approach to find explanations for performance differences was previously examined by Coff (1999) with his reaction to the differences between value creation and value appropriation. In his study, he specifically focussed on knowledge driven organizations. Knowledge driven organizations are those by which knowledge can be the source of a

competitive advantage (Barney, 1991). In these organizations, it is difficult to determine which stakeholder actually creates value. According to Coff (1999), individuals appropriate the actual value that is created; the organization on its own is not able to do this. Within an organization, the appropriation of created value depends on the bargaining power of each stakeholder (Coff, 1999).

Overconfidence and social value orientation are variables that influence human behaviour (Larkin & Leider (2012), Hvide (2002), Neale & Bazeman (1985)). Knowing that individuals are heterogeneous and that they create value collectively, (Felin & Foss (2005), Kahneman (1986), confirms that the social behaviour of an individual is important. Bridoux e.a. (2011) found that heterogeneous individuals have heterogeneous motivations. It is crucial that these different motivations are managed in the best way possible because different employees will eventually create collective value (Bridoux, 2011). One of the issues that can influence the motivation of the individual employee is whether the employee believes he is being treated fairly (Kahneman (1986), Stoelhorst & ElHaji (2012)). The use of fairness perceptions to analyse the microfoundations of the individual was used by Stoelhorst & ElHaji (2012). In their 2012 study, they investigated the differences in employee effort and fairness perceptions by looking at different firm situations. Fairness perceptions and the consideration of employees of being treated fairly are factors that were used by Kahneman e.a. (1986b) as well. Kahneman e.a. (1986) further explained that employees use reference transactions to determine fairness. This means that both the employee and company have a dual entitlement for appropriation of reference transactions. By combining this with the

(8)

knowledge that an optimal understanding and use of motivational systems should make companies more competitive, it is possible to conclude that fairness perceptions can be a tool to investigate overconfidence and social value orientations and its influence on the decision of whether a payment system is fair (Bridoux, 2011).

The importance of fairness perceptions will be discussed in more detail in the upcoming chapters. First, the current literature regarding overconfidence and social value orientation will be presented.

2.2 Overconfidence and social value orientation

Humans can be overconfident in their own abilities (Hvide 2002; Akerlof & Yellen 1990; Kruger & Dunning 1999; Squintani 2003; Santos-Pinto 2003; de la Rosa 2009). More & Healy (2008) described overconfidence in three ways. First, overconfidence can be the overestimation of an individual’s own actual performance. Second, it could be the overplacement of one’s performance relative to others. Finally, they mention the undue precision in one’s beliefs, . The authors concluded that difficulty with a particular task occurs when the overestimation of a person is higher and when the overplacement is lower.

One article that focusses on overplacement was written by Akerlof & Yellen (1990). This article emphasizes the fact that employees know that they all have differences in skills and performance, and that this affects productivity. These employees understand that being more productive means more compensation and they perceive this as being fair. Nevertheless, by evaluating themselves they are often overconfident and judge their productivity much higher than others would. Hvide (2002) indicated that this difference in perceived abilities is what is called bounded rationality from humans. He explains this phenomenon based on agents within an agent-principal setting. Agents typically emphasize the ‘good news’ about themselves and neglect the ‘bad news’ about their abilities. They want to present themselves positively to their supervisors. As a matter of fact, they are unaware of whether or not their confidence is realistic (Hvide, 2000). The author concluded that company management often feels obligated to offer higher wages to these overconfident employees. This does not appear to be logical however; it confirms the idea that a firm will provide different pay grades to socially heterogeneous employees. Hvide’s study did not focus on different payment systems.

(9)

A study by Neale & Bazerman (1985) was one of the first to pay attention to different incentive systems and social behaviour. Their research focussed on the level of employee confidence and their willingness to negotiate their pay. While playing a game that included bonus setting with an uncertain outcome, overconfident workers indicated that there would be more potential gains than more realistic workers did. This is an indication that different employees prefer different payment systems.

The combination of overconfidence and different payment systems was also addressed by Larkin&Leider (2012). They used an experiment to measure the confidence level of

employees. With this level in mind, they investigated what the most favourable incentive system would be for each subject. The results indicated that overconfident people tend to sort to a non-linear payment system instead of a linear payment system. Underconfident subjects were more likely to opt for a linear incentive system. This means that ‘underconfident’ employees and overconfident employees tend to choose different methods. The research by Larkin & Leider (2012) did not focus on whether employees thought these incentive schemes were fair.

Larkin & Leider (2012) did not mention any negative aspects of being overconfident; however, McGraw e.a. (2004) found results within his research that overconfident people more often feel less satisfaction with the outcome of a task. Because overconfident people thought more positively about their performance, they expected more than less confident players. According to McGraw e.a. (2004), this often results in a feeling of failure with a particular task.

An article by Fang & Moscarini (2005) showed that confidence level is directly linked to the morale of the employee. They found that different wage-models can affect the incentives for employees and it can also influence the morale in general. The employee moral and employee effort can be maintained at a high level if the correct wage-system is used. Especially when overestimation and overplacement occur (More & Healy, 2008), employee morale can be influenced with different payment systems. Fang & Moscarini (2005) concluded that using a variable wage-model is more risky than the non-differentiation wage-model.

De la Rosa (2009) approached overconfidence in a different way. Like Hvide (2002), he focussed on overconfidence and the difference in self-images in an agent-principal setting. In addition, De la Rosa (2009) recognized that the agent-principal setting is a good way to test the incentive contracts just as Hvide (2002) did. However, De la Rosa (2009) used a moral-hazard framework to test whether differences in overconfidence affect the higher or lower

(10)

power incentives. This indicates that with different incentive contracts, differences in confidence can be better managed by the principal. This study was conducted with CEO agents and their principals. It would be beneficial to conduct a more general investigation of employees and their behaviour in relation to different incentive contracts.

Instead of studying CEOs who are highly skilled workers, Kruger & Dunning (1999) focussed on un-skilled workers who typically have cognitive limitations regarding self-productivity. They concluded that people are imperfect in appraising their own abilities. They hypothesized that less skilled workers would significantly overestimate their abilities more than highly skilled workers would. However, after the study, they concluded that highly skilled people are poor at evaluating themselves. Important factors regarding overconfidence and biases in self-assessments are that there is often unclear feedback and evaluation from the authors.

Employees naturally attract positive feedback and deny negative feedback. As a result, the employee and the employer perceive the employee’s performance and productivity

differently.This difference in employee performance was examined in a tournament setting by Santos Pinto (2003). With different incentive systems, which were tournament prizes, workers were tested according to their risk-taking behaviour and their self-image within the game. They concluded that workers are good at spotting others who are overconfident, however they are unaware of the fact that they are evaluating their own productivity and performance higher than the employer would.

The tournament setting that Santos Pinto (2003) used in his research has also been used to find differences in social value orientations. These settings have been used to look for joint outcomes and equality outcomes in social behaviour (Van Lange, 1999). These outcomes should extend the assumptions of rational self-interest and indicate the non-rational behaviour of the individual that overrules the self-interest (Van Lange, 1999). Therefore, the assumption was made that individuals do not always only try to maximize their own outcomes, but also consider the outcomes of others. Kelley & Thibaut (1978) used the term ‘outcome

transformation’. These outcome considerations are connected to the payments systems used in for example game theory experiments (Van Lange e.a., 1997). These experiments search for differences in social value orientations and outcome decisions. In the experiments, a

distinction is made between pro-social and pro-self individuals. A pro-self individual is only concerned about himself and acts accordingly. Pro-social individuals are social individuals and are not only concerned about themselves, but also others. Van Lange e.a. (1999) found

(11)

evidence that pro-self or pro-social outcomes have important implications for individual human behaviour. Within an organization, this can affect the concern for colleagues, problem-solving behaviour between departments and it can influence common organizational goals (Nauta & de Dreu, 2002).

An article by Neale & Bazerman (1985) discussed one significant contextual factor for a company. They looked at the importance of the negotiator position in the behaviour of the employee. This means that in bargaining, it is important to notice the differences in

confidence-level considering that it affects the way the negotiator makes decisions about which contract to choose. Clearly, the overconfident negotiator opts for the arbitrary contract and confident, realistic negotiators will negotiate and devise a solution.

A notable remark within the research from Neale & Bazerman (1985) is the fact that

negotiators who discuss losses will more often choose the risky-arbitrary contract, while those who focus on gains will choose less risky negotiation contracts.

In general, a number of articles that have examined confidence focus on the risk-taking aspects of the effect of overconfidence. Neale & Bazerman (1985) as well as Dohmen et al. focussed on heterogeneity and how this is associated with people taking risks. In addition, theyfocussed on overconfidence. They stated that overconfidence influences the willingness to take risk. Despite the fact that this is a good indication in perceived behaviour, it does not explain whether employees think it is fair and it does not address what type of wage-system the employer uses.

2.3 Fairness perceptions

Thus, fairness, overconfidence and social value orientation are individual characteristics that are inter-connected (Van Dijk e.a., 2004; Fehr & Smidt, 1999; Santos Pinto, 2003). Burrows (1994) emphasized that both the individual and the company act in either a fair or an unfair way. He explained that unfair behaviour by the company would take advantage of abnormal conditions at the expense of other people. In bargaining situations, different circumstances occur and this is influences the perceptions of fairness. Evidence has been found that different levels of effort are put forth, despite the fact that ability and opportunities among employees are equal. This non-rational behaviour indicates that different aspects influence the effort employees put forth (Burrows, 1994). Burrows (1994) explained that not everyone is a team

(12)

player and that many people act selfishly. The degree to which the notion of fairness is recognised, affecting people’s attitude and behaviour. Therefore, it has been suggested that when modelling and analysing human cooperative behaviour, attitudes about fairness should be considered.

Kahneman e.a. (1986) previously questioned the importance of fairness perceptions in earlier research. In this study, they addressed the idea that the preferences for being treated fairly must be considered for not only the profit-maximizing firm, but also for individuals such as the employees and customers. The rational, selfish behaviour that was previously assumed to be a dominant trait is now accompanied by the common need for fair treatment. The

acknowledgement that fairness should be considered is a big step for profit-maximizing companies and could have great implications for explaining individual human behaviour. The working paper that Stoelhorst & El Haji published in 2008 is builds on the article by Kahneman e.a. (1986b). Kahneman e.a. (1986) had already indicated that an employee’s perception of the fairness affects the employee’s effort. In addition to that, Stoelhorst & El Haji (2012) focussed on environmental situations and their influence on actions taken by the firm of the employee. One of the factors that Stoelhorst & ElHaji (2008) tested was the firm’s profitability and the various actions the firm was executing at the time. They found that the

firm’s profitability, a contextual factor, affects the perceived fairness for employees. Different actions by the firm determine whether the employee thinks he is being treated fairly or

unfairly. Market forces and employment contract design are contextual factors that influence the employee. However, according to Stoelhorst & ElHaji (2008), there are no clear results regarding the differences in perceived fairness when either explicit or implicit contractual agreements are used. This means that no significant relationship was found between lower bargaining power and the contractual agreements.

Akerhof & Yellen (1990) discussed that firms generally take the expectations of the employee into consideration. Akerhof & Yellen (1990) suggested that this would decrease the employee effort because the employee may think the pay is unfair. Being paid fairy also implies that the employee is putting more effort into his job.

Cohn e.a. (2009) confirmed these assumptions and continued by making implications about the differences in payment methods. They concluded that if an employee is first paid unfairly, but when promoted and thus paid fairly according to the employee, that individual’s effort will increase. If an employee feels that he is first paid fairly and then promoted with no

(13)

change in pay, employee effort will remain the same (Cohn e.a., 2009). This experiment provides evidence that the fairness considerations clearly determine employee effort. The relationship between fairness perceptions and social value orientation was discussed by Fehr & Schmidt (1999). They acknowledged that firms (Kahneman e.a. 1986) and individuals could act in a pro-social way because of being treated fairly. The economic environment and the bargaining power of each stakeholder eventually determine whether pro-social or pro-self will be dominant (Fehr & Schmidt, 1999). This sorting effect, on which the employer does have an influence by managing it the best way possible, was also addressed by Hvide (2002), regarding the overconfidence of employees.

In addition, Van Dijk e.a. (2004) discussed how humans behave in either a social or a selfish way. One important implication is that the evidence showed that both the pro-social and the pro-self consider fairness perceptions and that they see it as some kind of bargaining power. Especially for observable behavioural, fair or unfair actions, selfish individuals tend to use and manage this (Van Dijk e.a. (2004)). If the behaviour is unobservable they do not mind being unfair. Individual pay and their perceptions are unobservable and not directly used as bargaining power.

2.4 Payment systems

Scientific studies about how to determine the most efficient company compensation schemes have been conducted for many years. One of the major issues regarding these decisions is whether to choose for either fixed payments or variable payments (Burke & Hsieh, 2005). This study focussed on the optimal point between these two payment systems and examined how this relates to optimal motivation and production. Typically, the fixed payment systems does not encourage extra effort or input for the risk-averse employee. A variable payment system is more suitable for risk-taking employees. Igalens e.a. (1999) attempted to frame all of the different options regarding payments systems. They divided the options into three categories; fixed payments, flexible payments and profit-sharing payments. With these options, conclusions were made about job motivation and job satisfaction. They found evidence that individuals are affected by these payment systems regarding job motivation and job satisfaction. Although these conclusions are not useful for this research, these different options can be helpful when trying to determine the fairest option according to employees.

(14)

One of the few studies that clearly concluded that different incentives result in different outputs was done by Lazear (2000). He found evidence that structuring of the incentive system differently can change the created value. He mentioned the importance of the

individual approach to the effectiveness of an incentive system. Lazear (2000) conducted his information at a company that was changing their payment system from fixed to variable (piece-rate payment). Both productivity and output changed with the change of the incentive system. Variable payment systems created more incentives for the employees en thereby changed productivity and output. Van Dijk e.a. (2000) discussed the variety of payments systems. Van Dijk e.a. (2000) tested whether there are significant differences between individual payments, team-based payments and performance based payments, with the focus on effort and motivation. Results showed that there were indeed differences in effort and outcome. For both individuals and teams, different motivations (competition, free-riding or cooperation) played a role. Effort thereby was affected by an employee’s perception of fairness.

Burke & Hsieh (2005) stated that there must be a balance between fixed and variable payments. They mentioned that risk-averse people would rather have a fixed compensation and that workers who are more confident would prefer a variable compensation. Nevertheless, they concluded that variable compensation is ultimately more profitable, and they

acknowledged that there must be a fit between fairness and satisfaction with the payment system.

Satisfaction and fairness of incentives are important for employees. This rationale was applied to research that attempted to discover which factors influence this perception (Dulebohn & Martocchio, 1998). This study investigated whether understanding of the pay plan,

satisfaction with base pay, organizational commitment, beliefs in the pay plan effectiveness, plan payout amount, and group identification influence the employee’s perception of whether he/she feels the salary is fair. Although organizational commitment and belief in the plan increase the chance of perceived justice, personal characteristics such as confidence and social value orientations are not mentioned.

To decide which payment system to choose for employees, it is important to look at the individuals that will formulate a perception of fairness concerning such a system. Considering the large difference in overconfidence between males and females, it is suggested that for fairness perceptions there is a significant difference in choosing different payment systems as well (Barber & Odean, 2001). These researchers conducted a study to find out whether there

(15)

are differences in overconfidence and risk aversion. Although risk aversion explains many aspects of overconfident people, it is not all about taking risks or not.

Using this information and the summary of the existing literature, the following hypotheses can be formed.

(16)

3. Hypotheses

This chapter will state the hypotheses that have been formed based on the literature that was previously discussed . These hypotheses are shown in a theoretical framework, which will explain the main relationships expected from the variables.

Akerlof & Yellen (1990) previously confirmed the influence of fairness considerations on the effort the employees puts in to their jobs. Stoelhorst & El Haji (2012) indicated that different environmental or economical situations create different fairness perceptions. Firms are forced to react to these situations if they want to be perceived as fair by their employees. Kahneman e.a. (1986) found evidence that customers consider and change their fairness perceptions if the environment is unstable. These environmental situations are external cases; firm does not have any influence on these circumstances. The payment system the firm chooses is not an external situation but an internal decision. Internal decisions and the consequences they have for an employee’s perception of fairness has not been investigated yet. There are many different ways to find an optimal compensation scheme (Igalens e.a., 1999; Burke & Hsieh, 2005). Which method is the most fair depends on the individual employee. As Van Lange e.a. (1999), Bridoux e.a. (2011) and Coff (1999) mentioned, the individual characteristics

determine human behaviour and decision-making.

According to Akerlof & Yellen (1990) and More & Healy (2008), overconfidence is related to overestimation and overplacement. If an employee were overestimating himself, he would expect more payment than he actually receives. In that case, a variable payment system would be preferred by this employee. In this system, he would be paid on a performance basis

instead of a fixed-basis.

Those who exhibit overplacement perceive their own value as higher than their colleagues, while this is incorrect. With these characteristics, an individual payment system would be preferred by this employee. Hereby, his performance is independent of his colleagues, unlike with a group compensation such as profit sharing (Igalens e.a., 1999). For this reason, the following hypothesis has been formed with respect to different individual, social

characteristics.

(17)

H1: ‘Overconfident employees perceive the individual payment method to be fairer than group payment, more often than confident employees’.

This hypothesis evolved from the assumption that overconfident employees think they create more value than their colleagues; which is overplacement (Akerlof&Yellen, 1990). They do not want to be dependent on the performance of their colleagues.

H2: ‘Overconfident employees perceive payment with variable compensation to be fairer than a payment through fixed compensation, more often than confident employees’.

This hypotheses suggests that overconfident employees consider variable compensation to be fairer than fixed compensation. They would prefer this method so they are sure to profit from the value they think they created. They will be compensated by the piece-ratio payment (Laezer, 2000), instead of a fixed payment.

The social value orientation was the second individual characteristic tested. The level of being pro-social or pro-self indicates different human behavior. (De Lange, 1999). As a result, the following hypotheses have been formed.

H3: ‘Employees who are more pro-social consider a group payment method to be fairer than individual payment, more often than pro-self employees.

H4: ‘Employees who are more pro-social consider a payment method with variable compensation to be fairer than payment with fixed compensation, more often than pro-self employees.

As with confidence, selfish employees tend to behave in a rational, self-maximizing way (Nauta & de Dreu, 2002; Van Lange, 1999). They do not want to be influenced and disturbed in their process of profit maximizing. Pro-social employees seem to perceive group payments as more fair because they consider organizational goals and commitment (Nauta & de Dreu; 2002). Pro-self employees do not want to be dependent on their colleagues; therefore, fixed payment is considered fairer than variable payment.

(18)

For the final hypotheses, the comparison between males and females is addressed. Males tend to be more confident than females and with this knowledge the following hypotheses have been formulated:

H5: ‘Male employees consider an individual payment method to be fairer than group payment, more often than female employees’.

H6: ‘Male employees consider payment with variable compensation to be fairer than payment with fixed compensation, more often than female employees.

(19)

4. Methodology

The methodology used in this research will be explained in this chapter. First, the indication of the sample population will be discussed. Thereafter, the research method will be explained and the research design will explain the vignette-study in more detail.

4.1 Sample

The research population consisted of people with work experience. Considering the fact that vignettes create a real-life experience, the participants needed to understand what it is like to be in the position depicted in the vignette. The participants who did not have any work experience were not included in the results. The respondents were randomly presented to either one of the vignettes. The vignettes will be explained in more detail later in this chapter. The research population (N= 199) were mostly from Northern America (75.9%). The

population was 56.8% male and 43.2% female and the average age was 35.2 years.

4.2 Method

The vignettes attempted to create a real-life company setting. Every individual that had work experience needed to show that they had experience dealing with the compensation aspect of their job. To test whether these employees or ex-employees had different thoughts about fair compensation, the fairness scoreboard from Kahneman e.a. (1986) was used. This 4 point-rating scale (Very Unfair, Unfair, Acceptable, Very Fair) attempts to indentify whether individual, social aspects of employees explain differences in fairness perceptions of a payment system.

Overconfidence has been calculated by a method from Adams (1957). He was one of the first to include confidence and overconfidence as a factor of influence on the social behaviour of individuals. The method used in this research will test whether the participant is actually confident in his/her own judgement. In this research, randomly chosen questions sought to find differences in confidence. By comparing the number of correct answers in percentages, with the percentage of confidence the participant had when filling in the correct answer, a

(20)

confidence-ratio was built. If they were very confident in filling in the correct answers, but they did not, they were rated overconfident. If they filled in answers that were more correct but evaluated themselves lower, they were seen as confident or insecure.

The other social aspect was calculated by the Social Value Orientation-measure used by Van Lange (1999). This measurement looks to find whether the participant is self or pro-social. The imaginary distribution between the participant and someone else can calculate whether an individual is selfish or social. The value is a comparison between the value you give yourself and others in a hypothetical game. The more you give to yourself relative to another, the more pro-self you are (Van Lange, 1999). This variable is important for both the third and fourth hypotheses.

4.3 Research Design

To test these hypotheses, a survey was created. Testing fairness perceptions is difficult to do with a regular survey considering the involvement of contextual and situational

characteristics. To study the considerations, different vignettes were created. These vignettes all depicted a different scenario. Considering all of the hypotheses, the vignettes describe either variable and fixed or individual and group payments. The following matrix (1) shows the structure of the vignettes.

Fixed compensation Variable Compensation Individual payment (1) Individual/Fixed (2) Individual/Variable

Group payment (3) Group/Fixed (4) Group/Variable

Matrix 1. Payment and compensation

1. Individual with fixed compensation:

You are currently working as a waiter/waitress. Your job is to serve the customers in a restaurant. Your compensation is fixed. The compensation is 10 Euro per hour. Tip money will not be part of your compensation.

(21)

You are currently working as a waiter/waitress. Your job is to serve the customers in a restaurant. Your compensation is variable. The compensation is 5 Euro per hour combined with a fixed percentage of the tip money you will receive from the customer.

3. Group with fixed compensation:

You are currently working as a waiter/waitress. Your job is to serve the customers in a restaurant. Your compensation is fixed. You will split 40 Euro per hour with 3 colleagues. The tip money will not be part of your compensation.

4. Group with variable compensation:

You are currently working as a waiter/waitress. Your job is to serve the customers in a restaurant. Your compensation is variable. The compensation is 5 Euro per hour combined with a fixed percentage of the tip money that you and 3 colleagues will receive from the customers.

These vignettes all represent a different payment system. In each vignette, most of the parts are the same, only the aspects that indicate the differences in payment system are different. Hereby, only the differences between the variables will be measured, which will increase the validity of the test. One of the strengths of the vignette study is the removal of the so-called self-serving bias (Stoelhorst & ElHaji, 2012). The vignette will simulate the situation the respondent should be in regarding the study, with no wrong or different interpretation. This is also one of the strengths of a vignette study. Konow (2003) used the vignettes study to test fairness studies, and acknowledged the fact that it is a flexible and easily controlled way to test the correct information.

These four vignettes all test the hypotheses proposed. To test these hypotheses, 2 vignettes were combined to see whether there was a difference in either group/individual or

fixed/variable. For example, if hypothesis 1 was tested, vignettes 1 and 3 were combined and vignettes 2 and 4 were also combined. For testing hypothesis 2, vignettes 1 and 2 were combined and 3 and 4 were also combined.

(22)

5. Results

In this chapter, the findings from the vignette-study will be discussed. First, some descriptive statistics will be listed, thereafter; a correlation matrix will be presented, concluding with a regression analyses of the hypotheses.

5.1 Descriptive Statistics

The research sample in this study included 199 respondents. In the sample, there were 113 males (56.8%) male and 86 females (43.2%). The average age was 35.18 years (SD= 11.062). 51 respondents had a part-time job (25.6%), 110 had a full-time job (55.3%) and 38 were currently unemployed (19.1%). More than half of the population had more than 10 years of working experience (55.8%). As mentioned previously, respondents who did not have any work experience were not included in the study because real-life experience regarding compensation schemes was necessary. Most of the respondents were living in Northern-America (75.4%) or Western Europe (14.6%). A majority of the respondents had completed a bachelor’s degree or higher education (56.8%). 28.1% of the respondents did not finish a professional degree.

Respondents were on average 25.54% overconfident (SD = 22.82). Of these respondents, 164 respondents were overconfident (82.4%). Only 35 respondents were found to be under-confident or under-confident (17.6%). These descriptive statistics are summarized in Table 1.

(23)

Table 1. Descriptive Statistics N Percentage Gender 113 56.8% Male Female 86 43.2% Ethnicity 150 75.4% 27 14.6% Northern America Western-Europe Other 22 20% Working Experience 9 4.5% 29 14.6% 50 25.1% 0 – 1 year 2 – 5 years 5 – 10 years 10 years or more ₁₁₁ _55.8% Current employment 51 25.6% 110 55.3% Part-time Fulltime Currently Unemployed 38 19.1% Finished Education 21 10.5% 92 46.2% 30 15.1%

Master or Doctoral Degree Bachelor Degree Technical Degree No professional Degree 56 28.1% Confidence-level 35 17.6% Under-or confident Overconfident 164 82.4%

The first findings from this research are the distribution from the different vignettes.

Kahneman e.a. (1986) and Stoelhorst & ElHaji (2012) previously showed how to present the different fairness perceptions from the vignettes. Differences in fairness perceptions prior to looking at the overconfidence and/or social value orientations is not one of the hypotheses, but can provide important information.

(24)

Vignette 1:

You are currently working as a waiter/waitress. Your job is to serve the customers in a restaurant. Your compensation is fixed. The compensation is 10 Euro per hour. Tip money will not be part of your

compensation. (Fixed, individual)

N=58

Fair: 63.8% Unfair: 36.2%

Very Fair: 17.2% Acceptable: 46.6% Unfair: 32.8% Very Unfair: 3.4%

Vignette 2:

You are currently working as a waiter/waitress. Your job is to serve the customers in a restaurant. Your compensation is variable. The compensation is 5 Euro per hour combined with a fixed percentage of the tip money you will receive from the customer.

(Variable, individual) N=45

Fair: 60.0% Unfair:40.0%

Very Fair:0% Acceptable:60.0% Unfair:31.1% Very Unfair:8.9%

Vignette 3:

You are currently working as a waiter/waitress. Your job is to serve the customers in a restaurant. Your compensation is fixed. You will split 40 Euro per hour with 3 colleagues. The tip

money will not be part of your compensation. (Fixed, group)

N=41

(25)

Vignette 4:

You are currently working as a waiter/waitress. Your job is to serve the customers in a restaurant. Your compensation is variable. The compensation is 5 Euro per hour combined with a fixed

percentage of the tip money you and 3 colleagues will receive from the customers. (Variable, group)

N=55

Very Fair: 0% Acceptable: 56.4% Unfair: 40.0% Very Unfair: 3.6%

5.2 Correlation

To find the correlation between the variables that were tested, a correlation matrix has been framed. All of the test variables have been included. The individual and group payment variable is set as a dummy variable with 1 = individual payment. The variable and fixed component as well as the gender variable are also set as a dummy variable in the analysis (1= variable payment; 1 = male).

There not a significant amount of correlation between the variables. Nevertheless, a

correlation does exist between the individual payment variable and the fairness perceptions r = .164 ( p = 0.021). A positive correlation exists between these variables. This means that individual payment is seen as more fair than group payment.

Other variables that indicate a certain degree of correlation are the work experience and gender. This negative correlation; -.165 (p = 0.020) indicates that within this sample, females had more work experience than males. In addition, a correlation exists between age and gender .233 (p = 0.002). In order to search for more influence, the gender variable will be investigated more thoroughly with the regression analysis. The final correlation between age

(26)

and working experience r= .590 (p= 0.000) can likely be explained by the fact that older respondents are more experienced employees. The complete correlation matrix is presented Table 2.

Table 2. Correlation Matrix

1 2 3 4 5 6 7 8 9 1. Fairness perception - - - - - - - - - 2. Individual Payment .164* - - - - - - - - 3. Variable Payment .031 .136 - - - - - - - 4. Overconfidence .057 -.034 -.029 - - - - - - 5. SVO-value .028 .010 -.076 -.103 - - - - - 6. Gender .105 -.010 .056 -.013 -.082 - - - - 7.Working Experience -.082 .124 -.010 -.004 .046 -.165* - - - 8.Ethnicity -.078 -.061 .056 -.086 .105 -.063 0.018 - - 9. Age .033 .109 -.046 .090 .069 -.233** .590** 0.017 -

** Significant at a 0.01 level (2-tailed) * Significant at a 0.05 level (2-tailed)

5.3 Regression

To test the hypotheses proposed in the theoretical framework, a regression analysis has been executed for each model. Starting with model 1, which questions whether overconfidence influences fairness perceptions with an individual payment, no significant results were found. This means that hypothesis 1 cannot be accepted. The variables included in the regression explain the fairness variable in the same way, but was not significant. The interaction variable; individual*overconfidence, is not significant to confirm the expected relationship. Model 2, which attempts to find an interaction between Variable Payment and

Overconfidence, was not proven to be significant (p > 0.05). As with model 1, the variables included in this model explain some of the effects of fairness perceptions, but are not

significant. Except for the insignificance, there are no variables that significantly explain the difference in fairness perceptions.

Model 3 is the first model that includes the SVO-value variables. Considering that the hypothesis suggests that pro-self employees will perceive individual payments as fairer, no significance was found. The interaction variable; Individual Payment*Pro-self, does not explain fairness perceptions as being significant (p = .197 > .05). Except for the insignificance in model 3, model 4 does not confirm the hypotheses. No significant explanation was found

(27)

with this model and no significant effect was found for the interaction variable; ‘Variable Payment*Pro-self’ (table 4.).

Models 5 and 6 attempt to find evidence of the effect of Gender on the fairness perceptions of the different payment systems. Model 5 combines gender with Individual Payment. For this interaction variable, no significant effect was found (table 4. p > 0.05). Except for the non-existence of a significant interaction variable, no variable was found that explained the fairness perceptions significantly. The last model in the regression attempted to explain the relationship between Variable Payment and Gender. For this hypothesis, there was also no significant result found (table 4.). The interaction variable; Variable Payment*Gender, did not have a significant effect on the fairness perceptions. In this model, no significant variable was found.

Table 3. Regression analysis Model 1-3

Model 1 Model 2 Model 3

Beta Sig. Beta Sig. Beta Sig.

Individual payment .094 .389 .082 .453 .074 .651 Variable Payment .012 .867 .082 .454 .087 .438 Age .125 .160 .127 .156 .134 .134 Overconfidence -.009 .925 .047 .696 .035 .745 Pro-self .041 .562 .040 .577 .057 .579 Gender (1= male) .119 .103 .129 .138 .111 .132 Working Experience -.147 .101 -.145 .103 -.146 .102 Individual*Overconfidence .119 .352 .135 .299 .153 .241 Variable*Overconfidence .112 .398 -.116 .379 Individual*Pro-self .198 .197 Variable*Pro-self Individual*Gender Variable*Gender R Square .066 .070 .078 F 1.691 1.581 1.595 Sig. .103 .123 .111

(28)

Table 4. Regression Analyse Model 4-6

Model 4 Model 5 Model 6

Beta Sig. Beta Sig. Beta Sig.

Individual payment .066 .687 -.053 .778 -.084 .657 Variable Payment .089 .586 -.090 .583 .039 .836 Age .142 .111 .140 .127 .137 .134 Overconfidence .031 .795 .031 .797 .018 .883 Pro-self .010 .934 .008 .947 .034 .790 Gender .108 .141 .118 .257 .203 .096 Working Experience -.152 .089 -150 .100 -.137 .134 Individual*Overconfidence .147 .257 .147 .258 .145 .266 Variable*Overconfidence -.097 .462 -.098 .460 -.099 .454 Individual*Pro-self -.185 .212 -.183 .221 -.210 .163 Variable*Pro-self .215 .154 .217 .154 .195 .201 Individual*Gender -.018 .894 .004 .974 Variable*Gender -.181 .182 R Square .088 .088 0.097 F 1.644 1.501 1.529 Sig. .089 .127 0.110

5.4 Other Variables

Although the hypotheses in this research could not be accepted because no significant results were found for the interaction variables in the regression analysis, useful information was discovered.

The correlations and the differences between the payment systems indicated differences of fairness. The correlation between the individual payment and the fairness perceptions has been investigated further (fig 1., table 5).

(29)

Table 5. Regression Analyses Individual*Variable Payment Individual preferences BETA Sig. Individual payment -.011 .986 Variable Payment -.175 .086 Individual*Variable .321 .013 R Square .058 F 3.971 Sig .009

Fig. 1. Fairness, Individual payment and variable payment

These statistics illustrate the differences in fairness for the individual payment system. The variable payment statistic is a conditional factor for the fairness of individual payments (p = 0.013). If there is a fixed payment in combination with the individual payment system (vignette 1), employees perceive the individual system as being less fair than when the individual payment system is combined with the variable payment system (vignette 2).

(30)

Hypothesis six (Model 6) ‘Male employees consider a payment method with variable compensation to be fairer than payment with fixed compensation, more often than female employees’, attempted to find the difference between males and females regarding their fairness perceptions of variable-based and fixed-based compensation methods.

With the regression, no significant difference was found. This means that no conclusion can be drawn from the assumption that male employees perceive variable compensation methods to be fairer than female employees. Nevertheless, a significant difference was found for gender which explains differences in fairness considerations. In conclusion, male employees perceive fixed payments to be significantly fairer than females (p=0.033) (Table 6; fig. 2).

Table 6. Regression analyses Gender

Gender preferences BETA Sig Variable payment .153 .157 Gender .213 .033 Variable*Gender -.206 .116 R Square .024 F 1.604 Sig .190

(31)

6. Discussion

This final chapter will summarise and discuss the findings from this research. Furthermore, the relevance of these findings in the academic field and recommendations for future research will be explained, followed by a general conclusion.

6.1 Summary of the findings

This research investigated whether there is a connection between individual social

characteristics and employees’ perceptions of fairness regarding the compensation scheme in a company. The participants in this study were divided into four groups; each group focused on one vignette. This selection was intentionally random so that no groups were selected on the basis of other variables. By combining the individual and group elements, and the fixed and variable elements, significant differences would confirm the hypotheses regarding the underlying preferences of the different individual employees.

This research confirms there are significant differences in fairness perceptions with different compensation schemes. Individual compensation schemes are seen as fairer than group payment schemes but only in combination with a variable payment system. This means that employees have a general preference for a payment system that is independent of other

employees but it also means that a condition for these preferences is variability in the payment system. A variable compensation scheme improves the likelihood of employees’ perception of fairness regarding an individual payment method.

Larkin & Leider (2012) previously found evidence that employees tend to prefer the same compensation schemes. In addition to preferring a payment system, this research confirms that employees already have a general preference for a payment system. This study

complements this by confirming that employees perceive individual payment methods as most fair.

This research also confirms that the microfoundations explained by Bridoux e.a. (2011) influence the decision about whether compensation is fair or not. Although individual characteristics such as overconfidence and pro-self behavior did not have a significant effect on fairness perceptions, gender was shown to actually explain differences in the fairness of a payment system.

(32)

6.2 Research implications

The first evidence that indicated the importance of the fairness perceptions from employees was found by Kahneman e.a. (1986b). Their view on value creation and value appropriation in the resource-based view, and the following individualistic view with Coff (1999), proved the importance of taking a close look at the individual employee. With knowledge concerning these employees’ perceptions of fairness in combination with different compensation

schemes, employee effort and reciprocity (Stoelhorst & ElHaji, 2012) can be stimulated. This is especially relevant with the knowledge that employees are trying to appropriate value that they think they deserved from the value creation (Coff, 1999). It is important to look at whether differences in confidence and social value orientation confirms a mismatch between the employees’ thoughts about the created value, and the thoughts of the employer.

Unfortunately this research did not confirm the relevance of social characteristics, such as the confidence-level and the social value orientation. It was assumed that these social, individual characteristics would significantly influence and explain the differences in fairness. Despite the fact that these characteristics did not confirm the hypotheses suggested, evidence was found of differences between payment systems and fairness.

Evidence was found in the literature of how employees react to differences in payment systems (Larkin & Leider, 2012), nevertheless there was no evidence found concerning fairness perceptions and these systems.

6.3 Limitations

There were potentially some fundamental limitations embedded in this research. Testing real-life events regarding an actual compensation scheme is not likely to be as secure as an actual real-life situation. To determine actual individual perceptions in the employer-employee situation and their actual behavior, specific interviews could be conducted instead of using a vignette-study.

The SVO-method is an accurate measurement of whether individuals are self or pro-social. This method implies that real money is distributed, yet it is still a fictional game. No real money is paid to the participants or others and thereby it is less accurate than real-life payouts.

(33)

The confidence-level was tested with an instrument that attempted to find differences in confidence by asking questions in random categories. This instrument builds an environment as real-life, but will never be the same as the actual social behavior of the employee.

Nevertheless, the instrument is an indication of the actual behavior of the individual. The research sample in this research is not a realistic basis for evaluating the complete labor force in the world. The respondents are mainly from Western Europe and therefore may not be reliable for other parts of the world.

The research-method from Adams that was used to determine confidence-level is outdated. Despite the fact that confidence-level can give an indication of the attitude of individuals, this research found no significant correlation with this variable.

6.4 Further Research

These results complement the theories from Stoelhorst & ElHaji (2012), Bridoux e.a. (2011), Coff (1999) and Kahneman e.a. (1986) regarding the influence of individual fairness

perceptions. Further research should focus on the individual employee and their preferences. Especially with sensitive subjects such as compensation schemes, considering fairness would be important. Therefore, more individual characteristics should be tested in combination with fairness perceptions. For example: promotion, international competitors, social responsibility and other aspects that can influence the employee’s perception of fair treatment. It is possible that these factors could influence fairness perceptions. In addition, the confidence level and social value orientation could have an influence on different the aspects of fairness

(34)

References

Adams, J. K., & Adams, P. A. (1961). Realism of confidence judgments. Psychological review, 68(1), 33.

Akerlof, G. A., J. L. Yellen. (1990) "The fair wage-effort hypothesis and unemployment." The

Quarterly Journal of Economics 105(2) 255-283.

Barber, B. M., & Odean, T. (2001). Boys will be boys: Gender, overconfidence, and common stock investment. Quarterly journal of Economics, 261-292.

Barney. J. (1991). Firm resources and sustained competitive advantage. Journal of Management. 17(1) Burke, L. A., & Hsieh, C. (2006). Optimizing fixed and variable compensation costs for employee productivity. International Journal of Productivity and performance management, 55(2), 155-162.

Bridoux. F, Coeurderoy. R, Durand. R, (2011). Heterogeneous motives and the collective creation of value. Academy of Management Review 36(4), 711–730.

Burrows. P, Loomes. G, (1994) The impact of fairness on bargaining behaviour. Empirical economics 19. 201-221.

Coff. R. (1999) When competitive advantage doesn’t lead to performance. The resource based-view and the stakeholder bargaining power. Organizational Science. 10(2). 119-133.

Cohn, A., Fehr, E., & Goette, L. (2009). Fairness and effort-evidence from a field experiment. Working paper.

De la Rosa, L. E. (2011) "Overconfidence and moral hazard." Games and Economic Behavior 73(2) 429-451.

Dulebohn, J. H., & Martocchio, J. J. (1998). Employee perceptions of the fairness of work group incentive pay plans. Journal of Management, 24(4), 469-488.

Fang, H., & Moscarini, G. (2005). Morale hazard. Journal of Monetary Economics, 52(4), 749-777. Fehr, E., & Schmidt, K. M. (1999). A theory of fairness, competition, and cooperation. Quarterly

journal of Economics, 817-868.

Felin, T., & Foss, N. J. (2005). Strategic organization: a field in search of micro-foundations. Strategic

Organization, 3(4), 441-445.

Hvide, H. K. "Pragmatic beliefs and overconfidence." Journal of Economic Behavior & Organization 48.1 (2002): 15-28.

Igalens, J., & Roussel, P. (1999). A study of the relationships between compensation package, work motivation and job satisfaction. Journal of Organizational Behavior, 20(7), 1003-1025.

Kahneman. D, Knetsch. J, Thaler. R, (1986) Fairness and the assumption of economics. The Journal of

Business 59(4). 285-300.

Konow, J. (2003). Which is the fairest one of all? A positive analysis of justice theories. Journal of

(35)

Kruger, J., & Dunning, D. (1999). Unskilled and unaware of it: how difficulties in recognizing one's own incompetence lead to inflated self-assessments. Journal of personality and social psychology,

77(6), 1121.

Lazear, E. P. (1996). Performance pay and productivity (No. w5672). National bureau of economic research.

Larkin, I., & Leider, S. (2012). Incentive schemes, sorting, and behavioral biases of employees: Experimental evidence. American Economic Journal: Microeconomics, 4(2), 184-214.

Larkin, I., Pierce, L., & Gino, F. (2012). The psychological costs of pay‐for‐performance: Implications for the strategic compensation of employees.Strategic Management Journal, 33(10), 1194-1214.

McGraw, A. P., Mellers, B. A. and Ritov, I. (2004), The affective costs of overconfidence. J. Behav. Decis. Making, 17: 281–295.

Moore, D. A., & Healy, P. J. (2008). The trouble with overconfidence. Psychological review, 115(2), 502.

Nauta, A., De Dreu, C. K., & Van Der Vaart, T. (2002). Social value orientation, organizational goal concerns and interdepartmental problem‐solving behavior. Journal of Organizational Behavior, 23(2), 199-213.

Neale, M. A., & Bazerman, M. H. (1985). The effects of framing and negotiator overconfidence on bargaining behaviors and outcomes. Academy of Management Journal, 28(1), 34-49.

Santos-Pinto, L., & Sobel, J. (2005). A model of positive self-image in subjective assessments.

American Economic Review, 1386-1402.

Stoelhorst. J, El Haji. A. (2008) Fairness considerations and value appropriation: exploring the conditions of human cooperation. Working paper.

Squintani, F. (2006). Mistaken self-perception and equilibrium. Economic Theory, 27(3), 615-641. Van Dijk. E, De Cremer. D, Handgraaf. M, (2004). Social value orientations and the strategic use of fairness in ultimatum bargaining. Journal of Experimental Social Psychology 40. 697–707

Van Dijk, F., Sonnemans, J., & Van Winden, F. (2001). Incentive systems in a real effort experiment. European Economic Review, 45(2), 187-214.

Van Lange, P. A., Agnew, C. R., Harinck, F., & Steemers, G. E. (1997). From game theory to real life: How social value orientation affects willingness to sacrifice in ongoing close relationships. Journal of

Personality and Social Psychology, 73(6), 1330.

Van Lange, P. A. (1999). The pursuit of joint outcomes and equality in outcomes: An integrative model of social value orientation. Journal of Personality and Social Psychology, 77(2), 337

Zelizer, V. A. (1996, September). Payments and social ties. In Sociological forum (Vol. 11, No. 3, pp. 481-495). Kluwer Academic Publishers-Plenum Publishers.

(36)

Appendix

Fairness perceptions

Thanks for participating in this survey. The answer will be used to get a better understanding of personal perceptions and human behaviour. The questionnaire is split up in 4 sections and will take you only 10 minutes to complete. All the answers will be anonymous and will be used only for research objectives. After filling in each section click on the arrow in the bottom right to continue. It is important to fill in all questions and click continue until no arrow appears.

Part 1. Vignette 1

You are currently working parttime as a waiter. Your job is to serve customers within a restaurant. Your compensation is fixed. Last week the compensation was 10 euro per hour. You didn't receive any tip money.

Vignette 2

You are currently working parttime as a waiter. Your job is to serve customers within a restaurant. Your compensation is variable. Last week the compensation was 5 euro per hour combined with a fixed percentage of the tip money you received from the customer. This weeks tip money was 5 euro per hour.

Vignette 3

You are currently working parttime as a waiter. Your job is to serve customer within a restaurant. Your compensation is fixed. Last week the compensation was split up along 3 colleagues, you earned 40 euro per hour. You didn't receive any tip money.

Vignette 4

You are currently working parttime as a waiter. Your job is to serve customer within a restaurant. Your compensation is variable. Last week the compensation was 5 euro per hour combined with a fixed percentage of the tip money you and 3 colleagues received from the customers.

Q1 To what extend do you consider the compensation scheme fair?  Very Unfair

 Unfair  Acceptable  Very fair

(37)

Part 2

In this task you have to imagine that you are randomly paired with another person, whom we will refer to as the other. This other person is someone you do not know and who will not know who you are. Your task is to allocate money between you and this other person. For each of the following questions, please indicate the allocation you prefer most by selecting the respective option at the bottom. You can only select one allocation for each question. After you have selected your option, enter the resulting allocation of money in the question below each allocation to confirm your choice.

In the example below, a person could chose to allocate money so that he/she receives 50 euros, while the anonymous other person receives 40 euros (Option 5).

There are no right or wrong answers, this is all about personal preferences. As you can see, your choices influence both the amount of money you receive as well as the amount of money the other receives.

Q2 Allocation 1 Note that the top row shows what you receive and the bottom row shows what the other receives.

Click to write Column 1 Option 1 (1) Option 2 (2) Option 3 (3) Option 4 (4) Option 5 (5) Option 6 (6) Option 7 (7) Option 8 (8) Option 9 (9) Which allocation do you prefer most?         

Q3 How much do you and the other receive in the allocation you chose? -You

-Other

(38)

-Other

(39)

-Other

Part 3.

The following questions are general questions of different topics. Try answering the questions and indicate your confidence in filling in the correct answer. If you don't know the answer and you are guessing, your confidence level is the lowest (50%). If you know the answer and you are absolutely sure, your confidence level is the highest (100%).

Q14 How many people are approximately living in Ecuador?  9.8 million people

 15.5 million people

Q15 How confident are you in filling in the correct answer? ______ Confidence percentage

Q16 The Netherlands aren't famous for their good weather. Although the amount of hours of sun is relatively high. How many hours of sun does the Netherlands on average have per year?  1200 to 1300 hours

 1500 to 1600 hours

Q18 Which artist got more number 1 hits in the United States?  Rihanna

 Madonna

(40)

Q20 The Netherlands aren't famous for their good weather. Although the amount of ours of sun is relatively high. How many ours of sun does the Netherlands on average have per year?  1200 to 1300 hours

 1500 to 1600 hours

Q22 Bill gates is still the largest shareholder of Microsoft. How many shares does Bill gates currently own?

 800 million  400 million

Q24 The cheetah is the fastest land animal. How fast can the cheetah run?  92 to 100 km/h (57 to 62 mph)

 112 to 120 km/h (70 to 75 mph)

Part 4.

Q26 What is your gender?  Male

 Female

Q27 Do you have a part-time or a full-time job?  Part-time

 Full-time

 Currently unemployed

Q28 How many years of working experience do you have (part-time and full-time)  None

 0-1 year  2-5 years  5-10 years  10 years or more