The effect of age on the relationship between incentives and job performance

(1)

The effect of Age on the relationship between

Incentives and Job performance

Name: Selcuk Kuzay Student number: 10646205

Thesis supervisor: dhr. prof. dr. V.S. (Victor) Maas Date: 19 July 2016

Word count: 12450

MSc Accountancy & Control, specialization Control

(2)

2

Statement of Originality

This document is written by student Selcuk Kuzay 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)

3

ABSTRACT

The aim of this paper is to investigate: a) the impact of use of financial and non-financial incentives on the employee job performance; and b) the moderation effect of age on the relationship between employee job performance and incentives. In total, 106 pairs of work floor employees/professionals and their supervisors/managers completed the survey. Operational employees are work floor employees or professionals, such as attorneys, consultants, doctors, engineers, salesmen, teachers and production line workers. This research is conducted in the Netherlands. Results revealed that financial incentives do not have an impact on employee job performance. Additionally, there is no relation found between non-financial incentives and employee job performance. Furthermore, age did not moderate the relationship between incentives and employee job performance.

Keywords: Employee job performance, financial incentives, non-financial incentives, age,

(4)

4

Employees are the most valuable asset of any organization (Furnham, 2005). Success of organizations lies in the hands of their employees. Employees will not perform to the standards and to the best of their ability unless they are motivated to do so. Employers, in both public and private sector, face problems when motivating their employees in order to improve performance. Organizations can increase the productivity if they know how to motivate their employees. Organizations seek for ways to create a motivational environment to be sure that employees’ work at their optimal levels to reach the objectives. To keep employees satisfied, many organizations use incentive systems to motivate and reward their employees. According to Bagraim et al. (2007), it is very important for an organization to determine the needs and goals of their employees in order to fulfil their needs and to attain the required motivation. When employees are motivated and satisfied, they become more productive and loyal to their work. However, when employees are not rewarded for putting extra effort, they will get dissatisfied over time and will not perform to the standards. This will lead to high employee turnover and poor performance, which in turn leads to lower profit (Westover and Taylor, 2010).

Incentives are one of the most used methods to increase employee motivation and performance. Ballentine et al. (2002) and Armstrong (2007) note that workplace motivators include both monetary and non-monetary incentives. Monetary rewards are also called financial rewards, and non-monetary rewards are called non-financial rewards. The monetary rewards include salary or wages, bonus, promotion, profit sharing and stock option, whereas non-monetary rewards include appreciation and recognition, job security, opportunity for growth and job enrichment (Hellrieger et al. 2004). According to Burgess and Ratto (2003), monetary rewards are used to stimulate a higher degree of employee-satisfaction, while non-monetary rewards, other than money, are used for employee-recognition, and provide a strong sense of security and stability. Employees will work harder to achieve a higher degree of recognition when they know that their job is secure and stable. Thus, recognition enhance the motivation of employees and leads to a better productivity (Heyman and Ariely, 2004). However, both type of rewards play a critical role in motivating the employees. Financial incentives have a short-term motivational effect on employees, while non-financial incentives have a long-term effect on the motivation of employees (Ellis and Pennington, 2004).

According to Popović et al. (2014), the effect of financial and non-financial rewards on employee performance is a continuous issue open for discussion and research. Previous research has shown that reward systems for employees are effective mechanisms for

(6)

6 stimulating performance and therefore, resulted in higher employee retention rates, productivity and job satisfaction (Cadsby et al., 2007; Culpepper, 2009; Heneman & Judge, 1999; Komaki et al., 1996; Stajkovic & Luthans, 1997, 2003). However, many studies focused on financial and non-financial motivational rewards on all levels of employment (Arnolds and Venter, 2007; Ramms, 2007; Chekwa et al., 2013; Igbaekemen, 2014; Amborse & Kullik, 1999). There are different levels of employees and they are motivated by different factors. Therefore, it is very important to determine which incentives have impact on employees at different stages. Moreover, needs and expectations of employees change rapidly. Employees could be either satisfied with monetary rewards or non-monetary rewards. Sarwar and Abugre (2013) note that there are still shortcomings in the way which supervisors can manage employee satisfaction and the knowledge of specific effects of rewards on job satisfaction.

According to Fullerton (1999), it is important to investigate whether the job performance for older employees is higher or lower as compared to younger employees. The average age of employees is increasing and projections estimate that, in 2050, the population of older workers will increase to 60 per cent (Carone and Costello, 2006). Age plays a crucial role in describing how employees change over time and influences the performance of employees (Avolio and Waldman, 1994). Kalleberg and Loscocco (1983) argue that studying age differences in job satisfaction among work is necessary as they lead to changes in role outcomes. Mather (2006) argues how ageing influences the reaction of employees to follow different incentive schemes, suggesting that older employees are more likely to be motivated by non-financial incentives while younger employees are satisfied with financial incentives.

The question then arises: to what extent is the level of relative financial and non-financial incentives associated with employee performance? The purpose of this research is to investigate the effects of financial and non-financial rewards on employees in line position in the lowest hierarchical level of an organization (operational employees) and the effect of age on employees in terms of incentives and job performance. The main contribution of this study is to extend the literature on employee performance and to explore new ideas on some variables that encourage employee performance. Moreover, this study has a different set up than previous research. Firstly, this study focusses on the performance of employees at operational level. Operational employees are work floor employees or professionals, such as attorney, consultant, doctor, engineer, salesman, teacher and production line worker. Secondly, this research is conducted in the Netherlands, while a lot of researchers conducted a research about this topic outside the Netherlands. Expectations and needs of employees can vary from country to country, such as physiological needs, safety needs, love and belonging, esteem,

(7)

self-7 actualization and self-transcendence, as described by Maslow (1943). Finally, survey-data was collected from pairs of operational employees and their direct supervisory managers in many different jobs and industries (Groen, 2012).

The objectives of this study are:

- To identify and determine the effects of financial incentives and non-financial incentives on employee performance;

- To identify and determine the effect of age on the relationship between incentives and employee job performance

This paper proceeds as follows. Section 2 provides the theoretical development of this study, including the hypothesis. In section 3, I will present the research methodology. In section 4, I address the study results. The last section describes conclusion and limitations.

(8)

8

2. Literature overview

Rewards, either financial or non-financial, have a major role in supporting the responsibilities among employees. These rewards ensure the coherence between performance and workforce (Wang, 2004). According to Samodien (2004), there is a lack of knowledge about which rewards effectively motivate employees. Armstrong (2000) note that a reward strategy, design of incentive systems, provides specific directions for organizations. These specific directions allow organizations to develop programmes ensuring that rewards have a positive outcome on employee performance. This means that when organizations use the type of rewards effectively, they can motivate individuals to perform at higher levels (Allen and Helms, 2001). Armstrong (2006) also mentions that a reward system will improve the motivation and commitment of employees, and increase job engagement. However, there are two types of incentives: financial incentives and non-financial incentives (Luthans, 1998; Armstrong, 2007). This study aims to investigate the prospects of financial and non-financial incentives undertaken by the management on operational employees.

2.1 The purpose of Incentives

Merchant and Van der Stede (2012) observed that even when employees understand what is expected of them, some do not perform as the organization expects because of motivational problems. Motivational problems are common because individual and organizational objectives do not naturally coincide (Merchant and Van der Stede, 2012). They argue that performance dependent rewards, or incentives, provide the impetus for the alignment of employees’ natural self-interest with the organization’s objectives. The use of incentives provide three main types of management control benefits: (1) informational; (2) motivational; and (3) attraction and retention of personnel. The informational benefit implies that incentives inform or remind employees how to direct their efforts in often-competing results areas, such as cost, quality, customer service, asset management and growth. The motivational benefit implies that some employees need incentives to perform tasks well, such as preparing paperwork and training employees. Employees who are pressured to do something are less likely to be willing to perform a certain task. By providing incentives, employees will perform these tasks. The third benefit, attraction and retention of personnel, involves that incentives are an important part of many employees total compensation package. Rewards, in this case, are offered to improve employee recruitment and retention by offering a market conform packages, and to employee’s continued employment. Another way to offer rewards is by providing

(9)

9 performance-dependent compensation elements in employees’ packages (Merchant and Van der Stede, 2013).

2.2 Financial incentives and Non-Financial incentives

The success of the use of financial and non-financial incentives to motivate employees has been examined by many researchers. There are many researchers who claim that financial incentives are more important in motivating the performance of employees (Al-Nsour, 2012; Baker et al., 1987; Burton, 2001; De Leiburne, 1991; Lawler, 1973; Naldöken et al., 2011), while other researchers are convinced that non-financial incentives have a higher effect in motivating the employees (Alwabel, 2005; Bragg, 2000; Bonner et al., 2000; Deci et al., 1981; Jenkins et al., 1998). Table 1 shows a range of different financial and non-financial incentives. However, according to Baker et al. (1988), employees are motivated by different factors. This means that one employee can be motivated by financial incentives, while the other is motivated by non-financial incentives.

Table 1: Financial and Non-Financial incentives

Financial incentives - Pay (Salary, wages and bonuses) - Pensions

- Insurances (health, life, accident, etc.)

- Allowances (travel, accommodation, clothing, childcare, etc.) - Subsidized meals/clothing/accommodation

- Subsidized transport

- Childcare subsidy/crèche provision Non-Financial incentives - Holiday/vacation

- Feedback

- Flexible working hours

- Access to/support for training and education - Sabbatical/study leave

- Planned career breaks

- Occupational health counselling - Recreational facilities

(10)

10 Financial incentives are considered as a short-term motivational tool on the performance of employees (Ellis and Penningtong, 2004). Jadallah (1997) concludes that financial incentives are used to enhance the recruitment of the capabilities of employees and their competencies level. Lawzi (1995) argues that with the use of financial incentives, to some extent, the employees will have control over their income. The employees will be evaluated based on their performance. How better they perform, how higher their income. Al-Nsour (2012) concludes in his research that there is a positive relationship between financial and non-financial incentives, and that the use of both incentives leads to a higher organizational performance. However, Al-Nsour (2012) finds that financial incentives have a higher impact on the performance than non-financial incentives. Burton (2001) and De Leiburne (1991) argues that financial incentives are an effective way to attract and retain staff.

Naldöken et al. (2011) conducted a research at a state hospital to determine the effects of the incentives on the motivation of employees. They find that financial incentives leads to a higher performance. Scheepers (2009) examined the use of financial and non-financial incentives at communication technology and information firms and the degree of the effect of the incentives. Scheepers (2009) concludes that employees are more likely to be motivated by financial incentives. Petty et al. (1992) examined the effects of financial and non-financial incentives at electric utility industry. In their case study, they report that 72 per cent of the employees within the business were satisfied with financial incentives and that the use of financial incentives reduced the turnover rates of the industry. Arnolds and Ventor (2007) examined the incentives which motivated blue-collar employees at manufacturing and clothing retail firms. They conclude that employees were mainly motivated by financial incentives. Pinar et al. (2008) conducted a similar research to 796 blue-collar employees. According to the findings of the research, employee productivity increases when financial incentives are higher. On the other hand, there are also many researchers concluding that non-financial incentives are more effective in motivating the employees, such as vacations and flexible working hours (Bragg, 2000). Ellis and Pennington (2004) report that non-financial incentives are considered as a long-term motivational tool on the performance of the employees. In addition, Armstrong et al. (2010) argues that providing a short-term motivational tool as a reward might result in long-term problems. In other words, when financial incentives are used as rewards rather than non-financial incentives, this may lead to future problems, which in turns leads to employee turnover. However, Gratton (2004) notes:

There have been many thousands of studies over the last decade that have looked at incentives and rewards. These have overwhelmingly found that while motivation is determined

(11)

11 by both monetary and non-monetary factors, money has come to play an overly important role in our thinking about the causes of behaviour. In most companies, very limited time and effort are spent on considering non-monetary sources of motivation. (p.126)

Jenkins et al. (1998) conclude that financial incentives do not always improve the performance of the employees. In addition, Deci et al. (1981) shows that financial incentives create other problems rather than enhancing performance. With the use of financial incentives, employees will tend to do their work quickly without having an eye on the quality of the work they perform. Tippet and Kluvers (2009) conducted a research about the use of non-financial rewards and their impact on performance. They considered non-financial rewards as a very effective instrument for enhancing job satisfaction and improving the performance. Kaya (2007) examined the use of financial and non-financial incentives affecting the performance of employees at hotel managements. Kaya (2007) found that non-financial incentives were more effective in motivating the employees than financial incentives.

In order to examine the changes of the views of employees, Towers Perrin (2003) surveyed more than 35,000 employees in U.S. companies. According to the study, financial incentives, such as competitive base pay and health care benefits, are the main drivers of attracting employees. However, non-financial incentives, such as career opportunities and development of skills, are important in retaining and engaging employees. This means that employees are higher motivated when non-financial incentives are available.

2.2.1 Expectancy theory

The expectancy theory is a process theory of motivation. The expectancy theory of Vrooms implies that the behaviour of an individual is instrumental; the individual tries to achieve something. The theory proposes also individual needs and motivation factors. Moreover, the theory implies that individual needs varies at different times and places (Vroom, 1964). Vroom (1964) also mentions that people specifically choose for behaviour, based on subjective expected outcomes and the idea that those outcomes are achievable and thus, result in higher motivation.

According to Vroom (1964), the expectancy theory has three variables that play a crucial role in motivating people. The first variable is the expectancy of employees. This variable concerns the relationship between effort and performance. In other words, the way how employees, in specific situations, assess that an effort will lead to the desired performance. The degree of the expectancy has a causal relationship with performance. The employees will be more motivated if the expectancy is higher. The second variable is the instrumentality. This

(12)

12 concerns the relationship between performance and rewards. This variable questions if a high performance leads to a positive outcome (reward) for an employee. The last variable is the valence. This concerns the attractiveness of the rewards and the degree to which an employee values specific rewards. The last two variables are positively related to motivation. When those variables have a higher degree, the employees will be motivated to perform well (Vroom, 1964). Figure 1 shows the basics of the expectancy theory.

Source: Thuis (2003)

According to the expectancy theory, employees will increase their effort: (1) when the probability of getting a desired outcome is high, (2) when the probability of getting a reward, cohesive with performing well, is high, and (3) when rewards are of high value. The theory implies that employees are more motivated by financial incentives. In addition, Brewer and Picus (2014) conclude that financial incentives are considered as the best motivational tool to motivate employees to achieve higher levels of performance when three conditions are met. First, the goals must be of value to employees. Second, employees must belief that goals supported by incentive systems are achievable and within their control. Finally, employees must perceive a relationship between effort and rewards (Brewer and Picus, 2014).

2.2.2 Two factor theory

The two factor theory, also known as ‘‘motivation-hygiene theory’’, proposes that there are two factors that causes satisfaction and dissatisfaction among employees, namely motivators and hygiene factors (Herzberg, 1957). Herzberg (1966) argues that these two factors are complement to each other. The motivators, also called satisfiers, serve as a source of satisfaction, such as achievement of targets, recognition, feeling that the employee is doing important tasks, responsibility, promotion and opportunities for growth. The hygiene factors, also called dissatisfiers, serve as a source of dissatisfaction, such as inefficient and ineffective

Figure 1. The expectancy theory model

Effort Performance Rewards

(13)

13 organizational policy, incompetent leadership or supervision, frustrating interpersonal relationship, poor work floor conditions and low salary (Herzberg, 1966).

Herzberg (1966) mentioned that satisfiers have a significant positive influence on job satisfaction, while dissatisfiers have a negative influence on job satisfaction. Satisfiers could lead to a small fraction of dissatisfaction when the level of satisfiers decreases or when there is a lack of satisfiers. In contrast, the elimination of dissatisfiers leads to short-time changes in the attitudes and does not lead to motivation and job satisfaction. Herzberg (1968) concluded that salary (money) serves as a dissatisfier when employees experience the salary they receive as low or as not sufficient. In other words, when financial incentives are low, employees will not be motivated to perform well. To motivate employees, motivation factors are strongly recommended. Hygiene factors are used to maintain the satisfaction of employees and making sure that employees does not become dissatisfied (Value Based Management, 2016).

According to Value Based Management (2016), a combination of the motivation and hygiene factors is possible and leads to four different scenarios. The first scenario implies high level of motivation and hygiene. When this is the case, the employees will be highly motivated. The second scenario implies low level motivation and high hygiene. In this case, the employees are not highly motivated and have few complaints. The third scenario implies high motivation and low hygiene. In this scenario, the employees are well motivated but they also have a lot of complaints. This is the case when employees experience the job as challenging, but they are not satisfied with the conditions and the salary they receive. The last scenario implies low level of motivation and hygiene. In this scenario, employees are unmotivated and have a lot of complaints. This is described as the worst situation (Value Based Management, 2016).

2.3 Theoretical framework

Figure 2. Conceptual Model: The influence of Financial and Non-Financial incentives on Job Performance.

Financial Incentives

Non-Financial incentives Job Performance

Gender Job tenure Educational level Age

(14)

14 This model describes the relationship of financial incentives and non-financial incentives with job performance, and the moderating effect of age between incentives and performance.

2.3.1 Incentives and Job Performance

Job Performance is a way to measure how employees are performing. According Jensen and Meckling (1992), managers are responsible for making sure that the interest of their employees aligns with organizational objectives. The managers can do this by using financial and non-financial incentives. This may provide huge benefits to any organization, since satisfied employees will produce more and stay reliable to the company. This positively influences the job performance of employees. According to Baker et al. (1987) and Lawler (1973), the use of incentives are considered as a very effective management tool for motivating and improving the performance of employees and job satisfaction. They believe that rewarding employees with incentives will motivate employees to perform well and to do exactly the tasks what they are told to do.

There are many researchers who claim that the use of incentives has a positive effect on job performance (Jenkins Jr., Mitra, Gupta and Shaw, 1998; Allen and Helms, 2001). The expectancy theory describes that employees will perform better when there is a probability of getting a reward, cohesive with performing well, is high (Brewer and Picus, 2014; Bonner and Sprinkle, 2002). In line with the expectancy theory, the two factor theory of Herzberg suggests that motivation factors (financial and non-financial) are compulsory in order to motivate employees and keeping them satisfied. This leads to a higher job performance (Value Based Management, 2016). According to Jenkins et al. (1998), using a reward system (financial and non-financial incentives) leads to favourable attitudes which in turn leads to higher job performance. In this study, job performance is defined as the extent to which employees meet their job requirements (Podsakoff and Mackenzie, 1989).

However, using a reward system does not always improve the job performance and may create problems rather that enhancing performance (Deci et al, 1981). Employees may tend to do their work quickly in order to be eligible for the rewards. This will lead to a decrease in the quality of the work they perform. According to Wiebel et al. (2010), financial incentives have a positive and a negative effect on performance. When employees experience a task as uninteresting, then it has a positive effect towards the performance. However, if an employee experiences a task as interesting, it has a negative effect towards the performance.

(15)

15 Contrary to the expectancy theory, the two factor theory proposes that when employees experience inefficient organizational policy, incompetent supervision and poor work floor conditions, they may become dissatisfied. This leads to a lower job performance. According to Gagné and Deci (2005), job performance is positively influenced when employees have the feeling that they have the choice to do a task, and is negatively influenced when employees are being pressured to do a specific task.

Hypothesis 1: Financial incentives have a positive effect on employee job performance. Hypothesis 2: Non-financial incentives have a positive effect on employee job performance.

2.3.2 Age and Job Performance

According to Kalleberg and Loscocco (1983), studying age differences in job satisfaction among work is necessary as they lead to changes in role outcomes. Avolio and Waldman (1994) argue that age plays a crucial role in describing how employees change over time, and that age influences the performance of employees over time. Employees have different socialization experiences and conceptions of what is desirable for work. This can be explained by age (Wright and Hamilton, 1978). This is in line with the expectancy theory. The theory implies that individual needs varies at different times and places (Vroom, 1964). Thus, when employees become older, their needs may change.

Previous research on the relationship between age and performance has shown mixed results. Baltes (1978) presented a theoretical framework, called ‘contextualism’, explaining how age is related to job performance. This framework is used to investigate how employee abilities and motivation may change during their life span. They conclude that the effects of increased age are twofold. Contextual factors may lead to a decrease of the performance of older employees. However, assigning new and challenging roles for older employees may retain or increase job performance.

Giniger, et al. (1983) argues that age is negatively associated with performance. They claim that an increase in age causes a downfall in the abilities of employees. Wright and Hamilton (1978) conclude that the increase in age may affect performance through motivation. Older employees are inclined to accept rewards that are available at organization level and lower their expectations, since they are going through a ‘grinding down’ stage.

According to Cleveland and Shore (1992), age explains a smart part of the total variance of declined job performance. They conclude that increased age does not affect the job performance in a direct way, since older employees have obtained more financial and

(16)

16 occupational resources compared to younger employees. Posthuma and Campion (2008), suggest that older employees are more satisfied, because they are resistant to any changes, more dependable, stable and honest. Rhodes (1983) argues that there is a positive relationship between age and non-financial rewards, implying that non-financial incentives increase with age. Rhodes (1983) also suggests that financial incentives decrease with age, suggesting that older employees become less motivated by financial incentives.

Hypothesis 3: The relationship between financial incentives and job performance is negatively

moderated by age.

Hypothesis 4: The relationship between non-financial incentives and job performance is

positively moderated by age.

The purpose of hypothesis 3 and 4 is to investigate the relationship between incentives and job performance. I expect that older employees are more likely to be motivated by non-financial incentives, while younger employees become more satisfied with non-financial incentives. The expectation of hypothesis 3 is drawn in figure 3a and the expectations of hypothesis 4 are drawn in figure 3b.

2.3.3 Control variables

The control variables used in this study are: gender, job tenure and educational level. According to Ali and Davies (2003), including employee variables may provide alternative explanations for employee job performance. They argue that gender plays a significant role in effecting performance. In addition to this, Nelson and Burke (2002) argue that the effect of gender differences is crucial in the work environment. Goldman et al. (2006) state that ratings of employees may be influenced by supervisors as result of gender stereotyping. Supervisors tend

P er fo rm an ce Financial Incentives Young Old Non-Financial Incentives P e rf o rm a n c e Young Old

Figure 3a: The relationship between financial incentives and job performance

Figure 3b: The relationship between non-financial incentives and job performance

(17)

17 to rate female employees lower compared to male employees. This leads to more favourable performance appraisals for men.

Roth et al. (2012) argue that differences between gender subgroups are an important issue when measuring job performance. Furthermore, Unger and Crawford (1992) perceive gender as a carrier variable for differences in experience and personal history. They argue that when the differences between male and female are controlled, the differences will disappear. According to Kanter (1977), the behavior of women should be more similar to the behavior of men, if they get the same opportunities.

In this study, I define job tenure as the length of employment in an organization in a specific position in the lower hierarchical level. According to Schmidt et al. (1986), job tenure and job performance are positively correlated to each other. They state that performance increases with experience. In addition, Judge (1994) argue that an employee may increase performance through higher levels of motivation. This motivation is increased by the alignment of an employee with organizational goals or by the employee’s person-organization fit, which can be achieved by the years of service that an employee provides. The more the years of service, the more the employee’s person-organization fit will increase. The human capital theory (Becker, 1964) also suggests that employees with a higher level of job tenure will perform better, since they have more job-related knowledge and more experience in the organization. In accordance with the human capital theory, the attraction-selection-attrition theory (Schneider et al., 1995) also suggests that person-organization fit increases with job tenure. Since the purpose of this study is to investigate the relationship between incentives and job performance, and how this relationship is affected by age, I decided to use job tenure as control variable.

The third control variable used in this study is educational level. According to Kasika (2015), educational level is a significant determinant in increasing job performance. The human capital theory (Becker, 1964) suggests that education influences the productivity of workers. The theory implies that higher educated people will perform better compared to lower educated people, since highly educated people will have more knowledge and skills compared to people with a lower degree of education. In addition, Kim and Mohtadi (1992) argue that education does not only have a positive effect on job performance, but also on economic development, economic growth and individual ability.

(18)

18

3. Research methodology

For this study, survey research approach is utilized. To the test the hypotheses, an online survey is conducted. The research study is conducted in the Netherlands.

According to Van der Stede et al. (2005), surveys should meet five criteria. The first criterion is that the researchers should have a clear and specific research objective for the guidance of the overall study. The purpose of this research is to investigate the effects of financial and non-financial rewards on employees in line position in the lowest hierarchical level of an organization and the level of rewards in the way that they are associated with performance. Furthermore, I will analyse and provide information about the factors that affects employee job performance. By doing so, I believe that this research will provide clear insights about the relationship between supervisors and operational employees, and between incentives and employee job performance. The second criterion is that the population should be defined and the sample taken from the population should also be clarified (see section 3.1). The third criterion is that the internal validity of the study should be questioned. The study should have a high internal validity. The fourth criterion is that the data should be accurate. The last criterion is that the researchers should report how they met the other requirements. The third, fourth and last criteria will be explained later on within this chapter.

3.1 Data collection and respondents

In order to get access to the database of the University of Amsterdam, which is provided by Bianca Groen, I had to find at least 6 pairs of respondents. A pair exists of an operational employee and his/her supervisor. Operational employees, in this case, are work floor employees or professionals, such as attorneys, consultants, doctors, engineers, salesmen, teachers and production line workers. The availability of data and the interesting research field with newly collected data, made the project attractable for me to participate.

In order to gather respondents, I started contacting people within my own network. If the people who wanted me to help were employees, I asked them to provide me the manager’s contact details. If the people who wanted me to help had a managerial background, I asked them to provide me the contact details of their employees. I also asked managers and employees if they could help me with finding other potential respondents. However, a snowball sampling is used in order to find potential respondents, since it is not easy to find respondents who meet the criteria. Snowball sampling, in this case, implies that potential respondents were asked for contact details of respondents who meet the criteria. Finally, survey-data was collected from

(19)

19 pairs of operational employees and their direct supervisory managers in many different jobs and industries (Groen, 2012).

The survey had to be filled in by both operational employees and their supervisors. Operational employees had to meet three criteria: (1) they are work floor employee/professionals who carry out the work; (2) they must have worked in their current function for at least one year; and (3) their supervisors had to use performance measures to measure the performance of their employees’. Supervisors/managers had to meet four criteria: (1) they had to be directly in charge of work floor employees/professionals; (2) they must have worked in their current function for at least one year; (3) they must have been in an operational department; and (4) they had to use individual and/or group performance indicators to measure their employees’ performance (Groen, 2012).

In total, 106 pairs of work floor employees/professionals and their supervisors/managers completed the survey, on a volunteer basis. According to Groen (2012), the low number of respondents is due to the fact that only few Dutch organizations have implemented a performance measurement system in their lower hierarchical level. However, two different databases were made. One database for the managers and the other for employees. The characteristics of the respondents are given in table 2. Since the purpose of this study is to investigate the relationship between incentives and job performance of employees, I will only use the surveys filled in by the employees.

Table 2: Respondent characteristics

Characteristics Employees

Gender 51% male

49% female

Education 16% Low-level high school

37% High-level high school 47% MSc or higher

Age Mean = 33 (SD = 9.2)

Job tenure Mean = 5.4 (SD = 5.8)

Size of the organization(#staff) Mean = 2,366 (SD = 8,606) min 1, max 70,000

Industry type employee 16% production firms

28% mass service firms

(20)

20

3.2 Survey instrument

The first version of the survey was tested among 17 respondents who had similar characteristics to the employees of the population. In order to shorten the survey and to check whether valid measures were used, three methods were used for the pretest: (1) Anderson and Gerbing’s (1991) item-sort task was used to determine how well the items measured the constructs; (2) Hak et al.’s (2008) Three-Step Test-Interview method was used to determine the respondents’ actual response behavior; and (3) the resulting survey data was analysed based on Cronbach’s alpha and principal component analyses (Groen, 2012).

The items that are used to measure the constructs of this study, in accordance with the main study of Groen (2012), are shown in Appendix A. Respondents, both professionals and their supervisors, had to rate all items on a 7-points Likert scale: (1) Totally disagree; (2) Disagree; (3) Moderately disagree; (4) Neutral; (5) Moderately agree; (6) Agree; and (7) Totally agree. However, respondents had the opportunity to fill in the survey in Dutch or English. By following the guidelines of Podsakoff et al. (2003), Bianca (2012) tried to prevent common method bias in several ways. The survey was constructed in such a way that each survey page only included the items regarding one construct that was introduced briefly. In this way, the quality of the data increased. This method helps the respondents to understand the items better (Frantom et al., 2002). The constructs are also measured in a different order than the order of the model, the confidentiality of the answers are emphasized and the scale items are pretested. By adding a latent common method factor to the model, common method bias is statistically controlled (Podsakoff et al., 2003).

The dependent variable in this study is employee job performance. This variable was measured with the scale for in-role job performance. The measurement of the variable is first developed by Williams (Williams and Anderson, 1991) and afterwards revised by Podsakoff and MacKenzie (1989). Employee job performance is a highly used variable in literature. This study uses the employee job performance to determine to extent to which managers perceive employees as meeting their job requirements (Groen, Wouters & Wilderom, 2015). Groen et al. (2015) further argues that the population exists of employees in different kind of jobs and industries. Thus, the scale is broadly applicable. Burney, Henle and Widener (2009) show that the scale correlates highly with objective measures of performance. The Cronbach’s alpha for this variable is .870.

Furthermore, the use of performance measurement for financial and non-financial is constructed for managerial needs. These two constructs are in line with each other, because both are used for the determination of performance rewards (Grafton, A., & Widener, 2010).

(21)

21 The independent variables used in this research are the performance indicators for incentive purposes, which is separated into financial and non-financial incentives. The independent variable is measured by Moers’ (2006) three subscales for using performance measure for incentive purposes. Despite this fact, in this study I only use financial and non-financial incentives. The items are measured on a 7-points Likert scale. Because of the items, I will run a factor analysis to determine what items are related to one factor loading. The Cronbach’s alpha for the ‘financial incentives’ variable is .820 and for the ‘non-financial incentives’ variable .796.

The moderating variable used in this research is age. Age is going to be measured from block 15 in the survey. In order to measure this variable, respondents will need to answer what their age is, which make this a scale variable. This variable is used by many previous studies and they suggested that age has an effect on financial and non-financial incentives associated with employee performance.

The control variables used in this research are gender, job tenure and educational level. The gender variable represents the males and females. At the end of the survey, in block 15, there are three questions which asks “What is your gender?”, ‘‘For how many years have you been working in your current position?’’ and “What is your highest completed education?’’. From these questions, the variables are going to be measured as the respondents will give information about their gender, job tenure and educational level.

3.3 Statistical analysis

The hypotheses are tested using Statistical Packages for Social Science (SPSS) version 22.0 to ensure the reliability and construct validity. The database consisted of 106 respondents in the first stage. The composition of the sample size is based on employees’ response to the related questionnaires. The respondents of this survey are all located in the Netherlands, which are using a performance measurement system in their lower hierarchical level (Groen, 2015). A multiple linear regression model is used to analyse the relationship between job performance and incentives. In order to test the reliability of the dependent variable, employee job performance, Cronbach’s alpha is used to determine the value. The Cronbach’s alpha for the dependent variable is .870. The reliability of the independent variables are also determined by Cronbach’s alpha. The Cronbach’s alpha for the independent variable ‘financial incentives’ is 0.820 and for the independent variable ‘non-financial’ incentives 0.796. Cronbach’s alpha values above 0.6 are considered to be acceptable (Nunnally, 1967). The Cronbach’s alpha

(22)

22 coefficient values for the dependent and independent variables are all above 0.6 and thus, I conclude that the items used in this study are reliable.

Before a regression analysis could be run, assumptions and preliminary analysis are required. The first assumption to run a regression analysis is by determining whether there is a multicollinearity issue by looking at the VIF values. The VIF values all should be less than 10. The independent and the moderator variables were centered before running the multicollinearity test. The VIF values of the variables are between 1.031 and 1.710. The values indicate that this assumption is met. The second assumption is the normality of the residuals. In order to test this assumption, scatter plots and histograms were created. The test shows a normal distribution and no outliers were found. The third assumption is to test the residuals for homoscedasticity. The Levene’s test is used in order to test this assumption. The Levene’s test resulted in an insignificant p-value of 0.523, which poses no issues with homoscedasticity. The last assumption of the linear regression is to look whether there is a statistical independence of the errors. The Durbin-Watson scale is used to determine whether this assumption is met. The Durbin-Watson value of the model resulted in a value of 1.506. This value should be between 1.5 and 3. Therefore, I conclude that all assumptions of the linear regression model are met.

The items in the questionnaire used in this study were tested on their validity. A factor analysis is an example of the preliminary test. To examine the adequacy for factor analysis, I used the Kaiser-Meyer-Olkin (KMO) and Bartlett’s Test of Sphericity. The KMO & Bartlett’s test is used to measure the sampling adequacy, to test the significance of the study, and to show the validity and suitability of the responses of the respondents collected. The KMO of this study is .808 and the Bartlett’s Test of Sphericity is .000. These two values imply that the factor analysis is suitable.

The factor extraction and factor rotation analysis were executed to determine the factor loading of the non-categorical variables and to interpret the questionnaire items. For the factor extraction, the Principal Component Analysis is used, and for the factor rotation, Varimax rotation is used. In order to decide how many factors will be used in this study, the scree plot of the eigenvalues and the component table is examined. However, the scree plot is used to determine how many factors will be used, since this criterion is more accurate than the eigenvalue-greater-than-1 criterion (Cattell, 1966). Based on the scree plot, two factors are used for the factor analysis. Factor rotation is used in order to interpret the pattern of loadings in an easier way. The results of the factor loadings and the descriptive statistics of the items can be found in table 3.

(23)

23

Table 3: Descriptive statistics and factor loadings of the measurement model

Variables α Factor N Min Max M SD

1 2

Employee job performance .870

Item 1 .856 -.097 106 1 7 5.73 1.047 Item 2 .862 -.042 106 1 7 5.84 1.015 Item 3 .800 .017 106 1 7 5.91 0.900 Item 4 .716 .112 106 1 7 5.34 1.294 Item 5 .766 .032 106 1 7 5.63 1.132 Item 6 .708 -.114 106 1 7 5.73 1.019 Item 7 .578 .087 106 1 7 5.57 1.219 Financial Incentives .820 Item 8 -.010 .880 106 1 7 4.54 1.790 Item 9 -.040 .869 106 1 7 4.52 1.948 Non-Financial Incentives .796 Item 10 .017 .912 106 1 7 4.57 1.927 Item 11 .039 .781 106 1 7 4.61 1.716 Eigenvalues 4.148 2.929 % of Variance 37.707 26.627

Table 3 shows that 64.334 percent is explained when two factors are used. As shown in the table, no questions are excluded when used two factors. However, when the model uses 3 factors, then some questions will be excluded. The model suggests that it is better to use the financial and non-financial independent variables as one independent variable. The Cronbach’s alpha, number of items, standard deviation and the mean are also given in table 3. From this table, it should be noted that the means of the items of the dependent variable is very high. Table 3 represents that all related items of the dependent variable have scored above .578. This means that the components representing the factor loading of the employee job performance and explains if they are correlating with each other. So with a value above the .578 none of the items have been removed for the usage. After a factor loading test, the Cronbach’s alpha of 0.870 have resulted, which has no need to be improved by eliminating levels. These variables could be used in further analysis because of the high reliability. Secondly, the financial incentive variable is also checked for factor loading and reliability. The items have shown a factor loading above .869. Afterwards, the Cronbach’s alpha is .820, which is still high enough for being reliable. For the items of non-financial incentives, the factor loading resulted in a minimum of .781. In addition, the Cronbach’s alpha resulted in .79, which is good and has no need to be improved. This means that employee job performance, financial and non-financial incentive variables could be used in further analysis.

(24)

24

3.4 Additional statistical analyses

In order to examine the independent variables, an additional factor loading analysis has been conducted. As seen in table 3, the factor analysis suggests that it is better to use the financial and non-financial independent variables as one independent variable against employee job performance. Table 3a represents the loadings of the financial and non-financial items.

Table 3a: Descriptive statistics and factor loadings of the independent variables

Variables α Factor N Minimum Maximum Mean SD

1 Financial Incentives .820 Item 8 .915 106 1 7 4.54 1.790 Item 9 .876 106 1 7 4.52 1.948 Non-Financial Incentives .796 Item 10 .863 106 1 7 4.57 1.927 Item 11 .793 106 1 7 4.61 1.716 Eigenvalue 2.979 % of Variance 74.485

The loadings of the items are between .793 and .915. This indicates that the factor strongly affects the variable. In other words, using financial and non-financial incentives as one variable would correlate better with the dependent variable. Furthermore, the use of financial incentives and non-financial incentives as a factor explains 74.485 percent of the total variance.

(25)

25

4. Results

In this chapter the results of hypotheses testing will be explained. In order to do test the hypothesis, data was stored and coded. The items in the questionnaire are all ‘positive’ and therefore, the items did not require any reverse coding. The total amount of the respondents was 106. After variable coding, the amount of respondents dropped to 102. Two respondents didn’t fill in their gender, one respondent didn’t fill in the question about job tenure and one respondent didn’t fill in the question about age. In total, there are four missing values. The regressions analysis is tested based on a sample size of 102.

Table 4 represents the correlation matrix of all variables, i.e. between the dependent, independent, moderating and control variables. The diagonal values represents the Spearman’s correlation, which is used to detect the linear relationship between two continues variables, whereas the Spearman correlation detects the relationship between ordinal and continuous variables (Podsakoff et al., 2003). The variables for this study exist of continuous and ordinal variables. Therefore, table 4 shows both the correlation types. As consistent with my expectations, it can be concluded that the relationship between financial and non-financial incentives is significantly positive (r = .775, p= .000). Financial incentives and age are negatively correlated to each other (r= -.050, p=.615), which is consistent with my expectation. However, I expected that non-financial incentives and age would be positively correlated to each other. Based on the outcomes, it can be concluded that age is negatively associated with non-financial incentives (r=-.122, p=.222). However, this correlation is weak. Furthermore, the relationship between job performance and financial incentives is positive (r=.110, p=.270). The relationship between job performance and non-financial incentives is also positive (r=.139, p=.163). I expected that the relationship between job performance and both financial and non-financial would be positively significant. It can be concluded that these variables have a weak positive relationship. The outcomes also show that gender is positively correlated to job performance, financial incentives and non-financial incentives. However, these correlations are weak (r= .134; r= .110; r= .035).

From table 4, it can be concluded that the independent variables have different types of relationships with the control variables. Financial incentives and education have a significant positive relationship with each other (r=.300, p=.002). The correlation between non-financial incentives and education is also positively significant (r=.208, p= .036). Job tenure is negatively correlated with both financial and non-financial incentives (r= -.079; r= -.175). It should be noted that job tenure has a weak negative correlation with job performance (r= -.195, p= .341), while I expected that these two variables would be positively correlated to each other.

(26)

26

Table 4: Correlation matrix: Pearson and Spearman Correlations

Variables 1 2 3 4 5 6 7 1. Job performance .098 .175 .023 .082 .062 .163 (.327) (.079) (.818) (.414) (.533) (.102) 2. Financial Incentives .110 .765** .030 .199* -.025 .107 (.270) (.000) (.762) (.045) (.802) (.286) 3. Non-Financial Incentives .139 .775** .001 .203* -.078 .043 (.163) (.000) (.991 (.041) (.437) (.665) 4. Age .008 -.050 -.122 -.020 .635** .012 (.938) (.615) (.222) (.839) (.000) (.907) 5. EducationA _.099 _.300** _.208* _-.126 _-.074 _.010 (.322) (.002) (.036) (.206) (.460) (.918) 6. Job tenureB _-.095 _-.079 _-.175 _.642** _-.172 _.001 (.341) (.432) (.078) (.000) (.083) (.992) 7. GenderC _.134 _.110 _.035 _-.005 _.000 _.075 (.178) (.269) _(.727) (.958) (1.00) (.456)

Notes: total n=102; **p<0.01; *p<0.05; The Pearson correlations appear below the diagonal; non-parametric Spearman correlations appear above the diagonal;

A_{1= primary education to 7= scientific education;} B_{job tenure in years;}

C _{0 = male, 1 = female.}

After the correlation matrix, all the required assumptions are met for running a multiple linear regression analysis. Because of the continuous variable, a multiple linear regression analysis would be the most suitable model. The multiple linear regression analysis is used to show the effects of independent variables on the dependent variable. The first step was to center the variables. This process implicates that the mean variance should be zero between the predictor variables, i.e. the independent and moderating variables. In this way, it was possible to compute the interaction terms and to decrease multicollinearity between the interaction term and the main effects. This means that the regression model is covering multicollinearity between the predictor variable and outcome variables. However, I could not find any multicollinearity between the variables since all VIF values are below 4. Multicollinearity would occur if the variables had a VIF value above 10. Furthermore, the multiple linear regression can be used when the assumptions are met. These assumptions are describes in statistical analysis (section 3.3).

Once the assumptions were met, I started to analyse the structural model. The regression analysis contains four models. The first model only includes the control variables, gender, job tenure and educational level, against employee job performance, the dependent variable. This model is used to determine whether there is a relationship between the variables. The second

(27)

27 model includes five variables, the three control variables and the two independent variables (financial and non-financial incentives). The second model is used to test the first and the second hypothesis. The moderating variable, age, is included in the third model. The interactions between the independent and moderating variables are shown in model four.

Table 5: Multiple linear regression analysis based on the unstandardized coefficient and

model fit of the structural model.

VariablesA _{Model 1} _{Model 2} _{Model 3} _{Model 4}

Independent variables: Financial Incentives -.017 -.019 -.023 (sig.) (.823) (.799) (.754) Non-Financial Incentives .063 .065 .070 (sig.) (.402) (.387) (.354) Moderators: Age .011 .009 (sig.) (.304) (.430) Interactions:

Financial Incentives x Age -.011

(sig.) (.124)

Non-Financial Incentives x Age .009

(sig.) (.198) Control variables: EducationB _.049 _.041 _.043 _.030 (sig.) (.410) (.515) (.499) (.635) Job tenureC _-.012 _-.010 _-.022 _-.015 (sig.) (.367) (.482) (.234) (.409) GenderD _.221 _.218 _.229 _.203 (sig.) (.159) (.171) (.150) (.203)

Model fit indices

R² .036 .047 .058 .082

Adjusted R² .006 .000 .000 .003

F 1.217 .953 .973 1.034

P .308 .450 .448 .416

*p< 0,10; **p< 0,05;

A_{Dependent variable: Employee job performance;} B_{1= primary education to 7= scientific education;} C_{job tenure in years;}

D _{0 = male, 1 = female.}

Table 5 shows the results of the multiple linear regression analysis based on the unstandardized B coefficient. Table 5 also shows the fit indices of R2, adjusted R2, the F-ratios and the p-values of the models. Model 1 has a R2 of .036 (3.6%), but increases in model 4 to .082 (8.2%). This implies that 8.2 per cent of the total variance in the dependent variable,

(28)

28 employee job performance, is explained by the use of financial and non-financial incentives moderated by the age of employees, including the three control variables (education, job tenure and gender). But a more adjusted ratio is the adjusted R², which shows the loss of predictive power in the regression analysis. In all the four models the adjusted R² resulted below the R². A possible explanation for this occurrence could be the high variability of the survey set. The F-ratios, commonly with the p-values, is used to test the fit of the regression analysis. However, it does not show any significance of the models.

Model 1 in table 5 contains the control variables with relation to the employee job performance. Model 1 tests whether there is control variables have an effect on the dependent variable. Education and gender have positive coefficients, while job tenure has a negative coefficient. The coefficient of education in model 1 implies that education has a weak positive relationship with employee job performance. Job tenure has a weak negative coefficient. This indicates that when an employee decides to work longer for an organization, the job performance will decrease each year. The coefficient of gender implies that females score higher than man. However, model 1 shows no significance between the dependent variable and control variables. The p-values are all above .05. Hence, no conclusions can be drawn from model 1.

Model 2 of the regression analysis includes the independent and control variables and tests the relationship between these variables and employee job performance. Model 2 is used to test hypotheses 1 and 2. There is no significance between employee job performance and financial incentives, and job performance and non-financial incentives. Financial incentives have a negative effect on employee job performance, while the coefficient of non-financial incentives is positive. This implies that higher financial incentives lead to lower employee job performance. In contrary, non-financial incentives increase employee job performance. However, these relations are not significant. Hypotheses 1 and 2 are not supported and therefore, the hypotheses are rejected. In addition, the null hypotheses for hypothesis 1 and 2 are supported. This is in line with the correlation matrix (table 4), which shows that there is no significant correlation between the dependent and independent variables. It should be noted that the R2 increases and the adjusted R2 decreases compared to model 1. If we add the independent variables to the model, the explanatory power of the regression model decreases, despite higher total variance explained. Hence, the model does not improve by adding the independent variables.

Model 3 contains the independent, moderating and control variables. Model 3 is also used to test hypotheses 1 and 2. The values of model 3 are comparable to the values of model

(29)

29 2, except that in model 3 the moderator variable is included. Age has a weak positive relationship with job performance implying that when an employee becomes older, the job performance of that employee will slightly increase. However, the relationship is not significant. Furthermore, adding the moderating variable does increase the R2 compared to model 2, but the adjusted R2_{does not increase. The model does not improve by adding the}

control variable.

Model 4 shows the moderating effect of age as an interaction between the use of incentives and employee job performance. Model 4 is used to test hypotheses 3 and 4. Based on the model, it can be concluded that there is no significant effect in the interaction terms. This means that hypotheses 3 and 4 are not supported and therefore, the hypotheses are rejected. However, the R2 and the adjusted R2 is higher compared to model 3.

It should be noted that, in the regression analysis, there is no significant relationship between the dependent variable and other variables. However, from the correlation matrix it could be concluded that the variables have a weak correlation with the dependent variable. None of the variables had a significant correlation with employee job performance. This indicates that the causal link between the variables is relatively weak. The multiple linear regression analysis is used to investigate the relationship of the dependent variable with the independent, control and moderating variable. Based on the outcomes of the regression analysis, there is no significant relationship between the dependent variable and other variables. Therefore, all hypotheses are rejected. It should be highlighted that the adjusted R2 is very low in all models, while the items resulted in having a high reliability based on Cronbach’s alpha and principal component analyses.

(30)

30

5. Discussion and conclusion

The main aim of this study was to identify and determine the impact of financial and non-financial incentives on employee job performance of operational employees. The second aim of this study was to explore whether age, as a moderating variable, had an impact on the relationship between employee job performance and financial incentives, and on the relationship between employee job performance and non-financial incentives. The research was conducted under broad range of Dutch-based firms, such as mass service firms, professional service firms and production firms. The theoretical framework of this study was drawn upon existing literature, in particular expectancy theory (Vroom, 1964) and motivation-hygiene theory (Herzberg, 1957).

For this study, I used a survey research approach. The survey had to be filled in by both operational employees and their supervisors. In total, 106 pairs of work floor employees/professionals and their supervisors/managers completed the survey. Since this study is based on employee job performance, the responses of the managers were not taken into consideration. The total number of respondents was 106. However, not every respondent filled in all the questions of the questionnaires. This resulted in a sample size of 102.

In order to test the aims of this study, four hypotheses were formed. The results of the study indicate that none of the hypotheses are supported. I couldn’t find any significant relationship between the dependent variable (employee job performance) and all of the other variables (financial and non-financial incentives, gender, educational level and age). However, I expected that the use of incentives would have an impact on the performance of employees as in accordance with prior research. According to Baker et al. (1987) and Lawler (1973), the use of incentives are considered as a very effective management tool for motivating and improving the performance of employees and job satisfaction. In addition, there are many researchers who claim that the use of incentives has a positive effect on job performance (Jenkins Jr., Mitra, Gupta and Shaw, 1998; Allen and Helms, 2001). Nevertheless, the results of my study reassert some findings consistent with prior studies. The correlation matrix (table 4) shows that there is a positive significant correlation between financial and non-financial incentives (r= .775, p= .000). This is in line with the findings of the study of Al-Nsour (2012) and Naldöken et al. (2011), as they suggested that financial and non-financial incentives strongly correlate to each other. Furthermore, no correlation was found between the dependent variable and the independent variables. This result is consistent with the study of Jenkins et al. (1998). In addition, Deci et al. (1981) also suggested that incentives do not always improve performance. They argue that financial incentives create other problems rather than enhancing

(31)

31 performance. With the use of incentives, employees will tend to do their work quickly without having an eye on the quality of the work they perform.

Cleveland and Shore (1992) argue that age only explains a smart part of the total variance of job performance. Considering the findings of Cleveland and Shore (1992), who conclude that increased age does not affect the job performance in a direct way since older employees have obtained more financial and occupational resources compared to younger employees, this finding might reflect the uncertainties that most employees are experiencing. From the correlation matrix (table 4), it can be concluded that age does not have a significant correlation with employee job performance, financial incentives and non-financial incentives. However, Rhodes (1983) argues that there is a positive relationship between age and non-financial rewards, and that non-financial incentives decrease with age, suggesting that older employees become less motivated by financial incentives. Nonetheless, the results of this study show that age does not affect job performance. It also shows that age does not have a link between the relationships of job performance with incentives.

5.1 Limitations

As with other quantitative studies, this research has some limitations. First, the survey results did not show any clear effect. This might be caused by the low sample size of respondents. Normality tests showed that there is moderate risk of sample representativeness. It is possible that the respondents do not represent the full population. Also the variable measurement could be the other explanation. Variables used in this questionnaire may not represent the constructs I aimed to investigate. Since the model did not reach a high fit, no conclusion can be drawn. Expanded models can include other antecedents to incentive systems (e.g. performance indicator, firm type, and firm performance). Second is the potential for high level of subjectivity, which can be caused by the set-up of respondents. Since both employee and his direct supervisor are invited to fill in the questionnaire, risk of ‘dressing up for visitors’ will appear strongly. Many professionals could answer in a desirable way, rather than their personal belief. This is more present when it comes to topic such as own performance, incentives and similar cases.

The last limitation of the research the reliability of the survey approach. Despite the high Cronbach’s Alpha of several variables, the variance explained in the model turned out to be very low. This might be caused by unequal distribution between the different firm types (production, professional service, industry) It would be interesting to see whether firm type also have influence on incentives and job performance.

(32)

32

5.2 Directions for further research

The results of this study provide some guidance for future research. This research focused on impact of incentives (both financial and non-financial) on job performance, considering age as predictor variable. Future research could examine the impact of different variables such as job satisfaction, firm size, compensation structure, strategy and organizational structure on the use of incentives. The use of incentives may differ depending on firm size or structure, which can lead to lower or higher job performance. Future research could also draw their study on other framework then expectancy theory (Vroom, 1964) and motivation-hygiene theory (Herzberg, 1957). This research showed that the predicted relations turned out to be not significant. More qualitative studies can be conducted to extend theoretical framework and to provide new insights into these topics.

The effect of age on the relationship between incentives and job performance

The effect of Age on the relationship between

Incentives and Job performance

Table of contents