Amsterdam Business School
The influence of outcome favourability on perceived fairness of
evaluation
Name: Mathieu Snoek Student number: 10871462
Thesis supervisor: dhr. prof. dr. V.S. Maas Date: 20 August 2017
Word count: 11004
MSc Accountancy & Control, both specializations
2
Statement of Originality
This document is written by student Mathieu Snoek, 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.
Abstract
A subject where there is still a lot of inconsistency in the results of research is the reliance on accounting performance measures and the effects it has on other subjects (Hartman, 2000). A specific part of this subject is the influence that outcome favourability has on the behaviour and attitude of employees. No previous research was conducted on how the outcome favourability of accounting performance measures influence the perceived fairness of an employee’s performance evaluation. My contribution to this subject will be an investigation of an employee’s perceived fairness as a result of a differencing outcome favourability. A lot of different factors have an influence on employees’ fairness perceptions. Aside from external factors, the actual outcome of a performance evaluation could affect this perception as well. Based on how people act with their own interest in mind (Lind & van den Bos, 2002) the argument that I make is that the perceived fairness of an employee’s performance evaluation is higher when they receive a more favourable outcome.
The outcome of the experiment showed that the favourability of an employees´ performance evaluation outcome has no significant impact on the perceived fairness of the received performance evaluation Further and deeper analysis of the perceived fairness perception did not show partial effects either.
4 Contents 1 Introduction ... 5 2 Theory ... 6 2.1 Fairness perception ... 6 2.2 Procedural justice ... 7 2.3 Distributive justice ... 8 2.4 Interactive justice ... 9 2.5 Performance evaluation ... 9
2.5.1 Accounting performance measures ... 11
3 Hypothesis development ... 13 4 Research methodology ... 15 4.1 Experimental design ... 15 4.2 Questionnaire items ... 16 5 Results ... 17 5.1 Demographic analysis ... 17 5.2 Independent variable ... 20 5.3 Dependent variables ... 20 5.3.1 Factor analysis ... 20 5.3.2 Reliability analysis ... 21 5.3.3 Normality assumption ... 22 5.3.4 Descriptive statistics ... 25 5.3.5 Correlation ... 27 5.3.6 Hypothesis test ... 28
6 Discussion and conclusion ... 32
References ... 34
Appendix 1: case plus questions in English ... 37
1 Introduction
In the first section of this thesis I will explain how I got to my subject and my research question. I will give a short review on the main concepts that helped me develop my research question. After reading this introduction, it should be clear in what way my thesis contributes to the existing literature.
For decades already, people work for firms and get rewarded (financially and otherwise) for the time, effort and knowledge they bring to the firm. As firms all over the world have implemented systems and rules to evaluated before mentioned performances by the employees to determine (a part of) their financial rewards, there has to be an upside to this way of working. Different research papers have concluded that this way of working can have big benefits for both employer and employee (Latham, Almost, Mann, & Moore, 2005). However, all systems and forthcoming rules have their disadvantages as well. In the case of performance evaluations, fairness is a big issue. Being treated fair has a big influence on employees, which, in the end, will affect the firm’s performance. As a part of fairness, favourability is a difficult subject to address. By not giving all their employees the same performance evaluation, firms reward employees with a good performance and vice versa. However, even though these evaluations can be explained as fair by the system, it can still have a big effect on the employees that have been evaluated (Lau et al. 2008).
There is still a lot of inconsistency in research on the effects that reliance on accounting performance measures for performance evaluations has on individual-level variables (Derfus, 2009). Fairness perception is one of these variables where it is unclear what effect accounting performance measures has on it. I want to contribute to this specific part of research by examining the reactions people have on the favourability of performance evaluations and the perception they will have on the fairness of the received performance evaluation. It could be that only less favourable results will affect the perceived fairness of the performance evaluation.
To contribute to this research I want to answer the following question: How does the
favourability of the outcome of an employee’s performance evaluation affect the perceived fairness of the received evaluation?
6
2 Theory
This chapter contains the theoretical constructs on which my research is built. First, fairness perceptions and how people make them is discussed. This includes several subsections on different types of fairness. Furthermore performance evaluations are discussed and what impact they have on firms and their employees. Finally, I discuss the how and why of reliance on accounting measures.
2.1 Fairness perception
Employees put in their time and effort into a firm and the firm then gives them a reward, mostly monetary, in return. This is the basic relation between employer and employee (Cropanzano & Mitchell, 2005). However, the real live version of the situation explained above is far too complicated to be able to capture it completely in just one sentence.
Firms use performance evaluations to value the work that has been done by an employee. For a firm to be successful in using performance evaluations to reach their goals, it is important that the performance evaluations of the employees are made in a way that they will improve the performance of a firm. Based on these performance evaluations, decisions will be made about an employee’s future. On the one hand improvement could be necessary, in which case appropriate measures need to be taken to make sure the employee’s performance will improve. On the other hand the performance measured can be great and a reward, monetary or otherwise, might be in place to keep the employee satisfied (Folger & Cropanzano, 1998). Lind & Van den Bos (2002) found that employees that are treated more fair will show behaviour that is desired by the firm, like accepting supervisors orders, obeying company policies and going “above and beyond” the expected effort. With fair behaviour of its employees, firms’ performances will rise. The opposite of this positive effect will happen when employees feel like they are treated unfair. Employees will start to act out of self-interest or in some cases even try to do damage to the firm even if it will not directly benefit their own interest.
As stated above, fair treatment of employees will influence a firm’s performance. Therefore it is important that the performance evaluation system of employees is seen as fair by the employees. Employees will react based on how they perceive the fairness of a performance evaluation system.
However, more recent research concludes that the perceived fairness of a system is measured based on different sub factors. These factors are procedural fairness, distributive fairness, interactional fairness, and voice. Procedural fairness looks at the procedures that are used to make the evaluation. Distributive fairness concerns whether the outcomes of the evaluations are perceived as fair. Interactional fairness refers to the interactions between employer and employee. Finally, voice reviews to if the evaluated employee was able to give his opinion and feels like he was being heard on the upcoming performance evaluation (Latham, Almost, Mann, & Moore, 2005). According to Lind & Tyler (1988) people often react more strongly to the fairness of the treatment they receive than to the favourability of the outcome. This means that the way a performance evaluation system is designed, being used and the way it presents the outcomes will have a big influence the type of behaviour employees are going to show.
2.2 Procedural justice
The field of research of procedural fairness examines the procedures that are used to make decisions when parties involved do not have the same interests. It looks at how people form a decision made based on the fairness of the process. Employees that feel that they are treated unfair, are likely to put in less effort and have a lower motivation.
Thibaut and Walker (1975) concluded in their research that the method used to come to a decision, the procedural fairness, has a big influence on the satisfaction and fairness. This finding comes with an important subsection. The use of a fair procedure can increase the satisfaction of all concerned, without any increase in the real outcomes available for distribution. This means that this finding is not affected by different types of circumstances and can be generalized to all kinds of evaluation processes.
Thibaut and Walker (1975) also found that people want to have an influence on a process if the outcome of that process is going to influence themselves. This is the case for both employer and employee during a performance evaluation. As employee and employer have different interests, the information being used in a performance evaluation should be gathered in the correct, predetermined way. By doing so, the employer ensures that an employee will feel like the process of obtaining information for his performance evaluation was conducted in a fair way.
8 relationship must be taken into account. Lind & Tyler’s (1988) discussion reviewed that procedural justice has an influence on a lot of factors that influence a long term relationship, something most people want out of a work relationship. Procedural fairness has an effect on job satisfaction, job performance, compliance with organizational rules, along with more key outcomes.
2.3 Distributive justice
Outcome fairness, or distributive justice focusses on the fairness of the outcome of a social exchange. Rawls (1971) defines distributive justice as the way to determine what people take to be the correct distribution of the available benefits and burdens of a social exchange, which includes the range of principals that people use to assign basic rights and duties
In case of an employee this will be how fair they perceive the outcome of their performance evaluation is. They compare the result of their received performance evaluation to what they think their peers deserves based on the effort everyone put in during the period which is evaluated.
People will use all the, for them available, information to determine if the way they have been treated has been fair. When there is no immediate information available about distributive justice, which is the case in most employees’ performance evaluations, they will turn to the information they do have at their disposal, which, in most cases, is procedural fairness information. A small company with only one employee in a certain role is a good example of not having distributive fairness information available, as there is no distribution of rewards of direct peers. So besides using procedural fairness information to access whether the procedure used was fair, they will also use the procedural fairness information to form an opinion on how fair their performance evaluation outcomes are (Van den Bos, Lind, Vermunt & Wilke, 1997). If employees feel that their work input and received performance evaluation outcome is not recognized fair compared to their peers, they will feel as being treated unfair. Being treated unfair will lead to all kinds of
The same effect occurs the other way around. This means that when outcome fairness information is available first, people will use this information to value their perceived fairness of the procedure they were part of.
2.4 Interactive justice
Greenberg (1993) sees two dimensions in interactive justice. Namely interpersonal justice, which revolves around the contact between employer and subordinate and informational justice, which focusses on availability of information.
Work by Bies (1985) found that interactive justice plays a big part in the fairness perceptions employees make of their performance evaluations. The main reason given is that employees can base their perception of interactive justice on all the interactions they have with the person who is responsible for their performance evaluation.
Bies & Moag (1986) identified four rules to be able to judge the fairness of interpersonal treatment:
Truthfulness. Authorities should be open, honest and candid in their communication when implementing decision-making procedures, and should avoid any sort of deception.
Justification. Authorities should provide adequate explanations of the outcomes of a decision-making process.
Respect. Authorities should treat individuals with sincerity and dignity, and refrain from deliberately being rude to others or attacking them.
Propriety. Authorities should refrain from making prejudicial statements or asking improper questions (e.g. questions that can lead to discrimination based on race, age, religion, etc.). By following these rules, authorities can make sure that they are being viewed as fair on an interpersonal level, which will have its effect on the perceived fairness of an employee’s performance evaluation.
Informational justice is classified as the content part of communication, the actual information that is being discussed. As much as this is general knowledge, people will have a higher fairness perception if the information they receive is correct and thorough. The same decision can be a lot easier to accept by an employee if the information on which the decision is based is shared. An example is that a raise of the budget cannot be given at the time due to financial restraints.
2.5 Performance evaluation
The relation between employer and employee is one of exchange. The employee gives his time, expertise and effort, whilst the employer rewards him for that with a payment of some
10 kind. In almost all cases this reward is (mostly) monetary. After the initial start of the relation between the two parties, events that occur during the relationship can cause it to change the dynamic of this relationship. An example of this is a promotion, including a raise, of a well performing employee.
Performance evaluations are used by firms to enhance the performance of the firm. This means that employees will get goals they can reach to get rewarded. These goals do have to be aimed at improving the firms’ results. However, besides trying to better the performance of the firm, goals and rewards are used to keep employees motivated and keep them with the firm. By making the goals and rewards part of the performance evaluation system, a balance needs to be found between the firms’ interest and the interest of its employees. This is where the fairness perception comes in to play.
A performance evaluation system can be classified as a management control system. To implement and maintain a system like that, the right environment needs to be created. A system needs to be helping employees instead of holding them back. So it is important for firms that they do not let bureaucracy get the upper hand in using systems. On the other hand, letting employees do whatever they want will lead to problems with the firms´ performance. This will happen because the goals of a firm and employee are not perfectly aligned and mostly even opposing.
For a fair process or procedure, Folger et al. (1992), identify three characteristics. The first one is adequate notice, which means that firms should publish, distribute and explain the content of the evaluation system to their employees to make sure they understand on what grounds they are judged.
Second is the fair hearing characteristic. A performance evaluation should include a formal meeting between employee and supervisor. In this meeting, the supervisor gives his assessment of the performance of the employee. After this the employee should be able to argue why he agrees or disagrees with the image the supervisor has of the work he has done. The third one is that all evaluations should be based on evidence. This seems like the only way, but there are a lot of cases in which other things, like personal relations or external pressure, are used as a basis to evaluate an employee. Favouritism should also be out of the evaluation, as all employees should be treated exactly the same.
Goal setting theory is also an important part as to why firms use performance evaluations. By setting a goal, firms can motivate employees to perform better. However, the success of goal
setting depends on three moderators. The first moderator is the importance of the expected outcome when a goal is reached. Employees will perform better when they give a higher value to the reward that is set for reaching the goal. Secondly, employees must feel like the goal that has been set is attainable. The final moderator is the commitment a specifically set goal creates. Promises to or deals with others can strongly enhance commitment (Locke & Latham, 2002).
Lau and Sholihin (2005) examined the effects of financial and non-financial performance measures on job satisfaction of employees. Their findings align with the goal setting theory, as they found that the presence or absence of performance measures influences employee job satisfaction. These findings provide proof of a complicated link between performance measure specifications, fairness perceptions of employees, trust in supervisor and job satisfaction level.
Merchant (1990) conducted a survey among managers and asked directly for actions that were not in the best interest of the company but in the self-interest of the manager. The majority of these managers admitted that they had intentionally engaged in such activities and blamed the pressure to meet financial targets and high environmental uncertainty as reasons. Many other papers relate to CEO compensation, and evidence of manipulation is found repeatedly. Within the bounds of law, other examples of changes in accounting procedures and impaired operating decisions include studies by Holthausen et al. (1995), Gaver et al. (1995), and Healy (1985). Clearly illegal ways of manipulation through misstatements of earnings or financial statements were found by Efendi et al. (2007), Burns & Kedia (2006), and Bergstresser & Philippon (2005). As mentioned by Goldman & Slezak (2006), the financial scandals of Enron, WorldCom and others provide additional evidence on manipulative behavior which aimed at personal wealth increase. In the course of that, the scandals caused increased public as well as political attention regarding the issue.
2.5.1 Accounting performance measures
Over the years there has been a lot of criticism on the reliance on accounting performance measures for employees’ performance evaluations (Hartman, 2000). However, to be able to measure performance objectively, numbers that employees cannot alter in any other way than by just doing their job they are supposed to do are still needed. With the way accounting information has to be collected by law and regulations, using those numbers is the best suitable option.
12 By using accounting information for performance evaluations, other relevant aspects of an employee’s performance are neglected. Accounting numbers only show the outcome of the work done and do not give any indication of the process how the outcomes were reached. The same applies to external factors that could have an influence on the accounting information used (Hopwood, 1972). Another negative effect of using accounting performance measures to evaluate employees are that it is almost impossible for a firm to exactly measure its economic costs function. This makes the reliance on accounting performance measures less reliable. Opposed from this criticism, accounting performance measures has upsides as well. The use of accounting performance measures leads to better performances of employees (Lau et al., 1995). As discussed earlier in this chapter, procedural fairness of a performance evaluation is an important issue for firms to make sure the wanted effect is reached (Chenhall, 2003). In the end, enhancing firm performance is an important reason to be using accounting performance measures. And if that can be reached with a favourable trade-off compared to other factors, most firms will apply it.
3 Hypothesis development
In this chapter I will describe how I developed the hypothesis for my research. I will draw from the theory that I described in the previous chapter. It should be clear what my research will comprehend and how this adds to the existing literature.
If all the employees are evaluated the same way, the outcome of these performance evaluations shouldn’t affect the perceived fairness of outcome of the employee’s performance evaluation, no matter how favourable the performance evaluation is for them. However, employees with a more favourable evaluation will probably rate the evaluation process as more fair compared to employees with a less favourable evaluation. Subconsciously the favourability of a performance evaluation will have an effect on the perceived fairness of the outcome, even when the performance evaluation process was exactly the same, and therefore fair, for everyone (Lau et al, 2008).
The distribution of the outcomes of performance evaluations should be seen as fair by all employees, as they are all assessed and evaluated in the same way. In a perfect world there will be no difference between the favourability of the performance evaluation outcome and the perceived fairness of the performance evaluation outcome. However, we do not live in a perfect world and base our decisions on other things than the aim of perfection and equality for everyone.
The main reason for a different effect compared to the perfect situation is the self-interest all parties will have in a social exchange. Having an actual better and more favourable outcome will lead to a higher personal satisfaction and that will be the reason that evaluated employees with a more favourable result will perceive the received outcome as more fair (Lind & Van den Bos, 2002).
People show more positive behaviour and have a more positive attitude if they find they have been treated fair. When the outcomes of their performance evaluations are more favourable, they perceive the process to be more fair compared to people with a less favourable performance evaluation outcome (Skitka, Winquist, & Hutchinson, 2003).
Outcome information, in this case the favourability of it, is used by people in a self-serving manner. They will take credit themselves when a performance evaluation outcome is favourable for them, but when a received performance evaluation outcome is less favourable for them, people will try and ascribe this to other factors. This self-serving bias makes that
14 people with a more favourable performance evaluation outcome will perceive the process of this evaluation as more fair compared to people with a less favourable performance evaluation outcome (Francis-Gladney et al., 2010).
As a result of this, I expect that a performance evaluation outcome with a less favourable outcome will be considered as less fair compared to a more favourable performance evaluation outcome. My hypothesis for this is:
Employees with a less favourable performance evaluation outcome will have a lower perceived fairness than employees with a more favourable performance evaluation outcome.
4 Research methodology
In order to test whether my hypotheses are confirmed in practice, I have performed an experiment. In this chapter I will explain my research methods. First, I will discuss the design of my experimental case. Secondly, I will explain the variables that I plan on using.
4.1 Experimental design
The most appropriate method to use for my thesis project would be an experiment. It would be very difficult to use actual data because all others factors that can impact the employee fairness perceptions would have to be held constant. The benefit of using experiments is that a high internal validity can be achieved. (Schulz, 1999; Smith, 2014). For experiments in management accounting it is feasible to use students (Liyanarachchi, 2007). A total of 50 respondents should be enough, because of the high internal validity of experimental research. I got the respondents by asking fellow students, business contacts and friends and family to participate. I sent everybody a personal message asking them if they want to help with my thesis by filling in a questionnaire about performance evaluation. All participants completed the questionnaire digitally (smartphone or computer). I have translated the survey into Dutch to make it easier for Dutch speakers to understand the case and questions.
The case is about a sales professional in the sales department of in a company. It is stated that the company is a trading company in the machine parts industry in The Netherlands. Sales employees have a lot of influence on the sales they do. The case mentions the performance of said sales professional and it will be required to answer questions based on the story and information given.
The performance of sales employee Marc in the case will be divided in two groups that will each get a different version of the case. The first group gets a case where Marc receives an unfavourable performance evaluation. In the second group, the sales Marc generated will result into a favourable performance.
Conditions with a favourable performance evaluation of Marc:
This year Marc generated € 425.000 in revenue. With this revenue his evaluation has a B value. Because of this rating, Marc will receive a performance bonus of 5% of his salary.
16 Conditions with a unfavourable performance evaluation of Marc:
Last year Marc generated € 325.000 in revenue. With this revenue his evaluation has a D value. Because of this rating, Marc is not qualified for a performance bonus. He also has to follow an improvement plan to improve his performance in the coming year.
4.2 Questionnaire items
The operationalization of the variables that I use in my experiment will be explained in this paragraph. In appendix 1, I included the English version of the questionnaire and in appendix 2 the questionnaire that was translated to Dutch.
In order to measure fairness perception as a dependent variable, I asked three questions. These questions will show the opinion of a respondent on the perceived procedural fairness of the case, the perceived outcome fairness and the perceived fairness of the case as a whole. Before the manipulation as discussed in paragraph 4.1 is shown, I first ask the respondents to answer three general questions that show their stance on performance evaluations in general. To check whether or not the respondent paid attention when reading the case, I ask how favourable they find the received performance evaluation. By doing so I also make sure that the respondents perception of the performance evaluation outcome aligns with the perceived perception of the case.
To conclude the questionnaire, I ask questions about the respondent self. These questions will give me information about age, gender, education level, work experience, and experience with performance evaluations.
5 Results
An overview of the results of the conducted experimented will be presented in this chapter. First, the analysis of the demographics of participants is discussed. Next are the independent and dependent variables. Finally, the hypotheses are tested. Below is an overview table of the names of the variables that are used.
Table 1: Variable names
EvaluationOutcome Nominal variable: 0 = negative outcome, 1 = positive outcome
Education Ordinal variable: 1 = University, 2 = Professional, 3 = Intermediate, 4 = High School, 5 = Other
ResponseTime Nominal variable: 0 = first day, 1 = later Language Nominal variable: 0 = English, 1 = Dutch Age Continuous variable: number of years Gender Nominal variable: 0 = man, 1 = woman
WorkExperience Continuous variable: number of years of working experience
EvaluationExperience Continuous variable: number of years of experience being evaluated based on work performance
EvaluationAttitude Continuous variable: measured on a 7-point Likert scale FairnessCombined Continuous variable: measured on a 7-point Likert scale ProceduralFairness Continuous variable: measured on a 7-point Likert scale OutcomeFairness Continuous variable: measured on a 7-point Likert scale TotalFairness Continuous variable: measured on a 7-point Likert scale
5.1 Demographic analysis
As most of my family and friends are Dutch and will have a better understanding from a Dutch questionnaire, I created both a English and Dutch version of the questionnaire. Of the 53 participants, only 2 filled out the questionnaire in the English language, whilst the other 51 answered the questionnaire in Dutch. After performing an Independent Samples T-test with Evaluation Outcome as dependant variable, it can be concluded that there were no significant
18 differences in the distribution between the questionnaires’ languages (t-statistic of 1,445 and a p-value of 0,161. Spearman correlation = -0,210; p-value 0,132).
In the next table, an overview of the demographics between both groups is presented. The average age in the positive outcome the average age was 32,54 and for the group with the negative outcome this was 37,56. Of the 53 valid responses, 37 questionnaires were filled out by men and 16 by women. Across both groups the average work experience was years. And the average number of years of experience being evaluated on work performance was.
Table 2: Descriptive statistics of the two groups
Evaluation outcome Negative(n=25) Positive(n=28) Age Mean 37,56 32,54 Std. Deviation 15,226 12,808 Std. Error Mean 3,045 2,420 Gender Number of men 17 20 Number of women 8 8 Mean 1,32 1,29 Std. Deviation 0,476 0,460 Std. Error Mean 0,095 0,087 Work experience Mean 18,04 14,61 Std. Deviation 13,306 10,668 Std. Error Mean 2,661 2,016 Evaluation experience Mean 9,12 7,21 Std. Deviation 9,584 7,031 Std. Error Mean 1,917 1,329
Even though the questionnaire was mostly distributed amongst friends and relatives, most of the respondents had a professional or university education. One third of the respondents has an education level lower than this. The Dutch population is getting more and higher educated
in general. Plus that higher educated people generally receive more extended work evaluations, this seems like a good reflection of the population to me.
Table 3: Highest completed education
Negative(n=25) Positive (n=28)
Frequency Percentage Frequency Percentage
University 8 32% 7 25% HBO 8 32% 12 43% MBO 6 24% 6 21% High School 3 12% 2 7% Other 0 0% 1 4% Total 25 100% 28 100%
An Independent Samples T-test with the independent variable of Evaluation Outcome is shown in table 4. This test is done to find potential demographical differences between both questionnaire groups. Even though Levene´s test for homogeneity of variance is significant for age and work experience, there is no statistically significant difference between the age of both questionnaire groups in the Independent Sample T-test (t = 1,292; p = 0,203). The same goes for the years of work experience (t = 1,028; p =0,309) both of those. As for gender, no differences were found between both groups (t = 0,266; p = 0,791). The difference in level of education between groups is not significant (t = -0,192; p = 0,849). Finally, for evaluation experience there is no significant difference between both groups either (t = 0,831; p = 0,410).
20
Table 4: Independent Samples T-Test (with Evaluation Outcome as independent variable)
Age Gender Education WorkExp EvaluationExp
Levene's test (sig.) 0,012 0,599 0,747 0,027 0,137
T statistic 1,292 0,266 -0,192 1,028 0,831 Degrees of freedom 47,168 51 51 45,989 51 Significance (2-tailed) 0,203 0,791 0,849 0,309 0,410 Mean Difference 5,024 0,034 -0,054 3,433 1,906 Std. Error Difference 3,890 0,129 0,283 3,339 2,292 Lower 95% confidence interval -2,801 -0,224 -0,623 -3,287 -2,696 Upper 95% confidence interval 12,849 0,293 0,515 10,153 6,508 5.2 Independent variable
For the manipulation check I used a question to determine if they read and understand the case. By asking how favourable the performance evaluation of Marc was I tried to make sure that besides just reading the case, that people understood the situation Marc was in and are able to answer the questions according to the manipulated scene created for them. I got 58 completely filled out questionnaires. Of these 58, 5 of them did not overcome the manipulation check as they filled out an opposing view compared to the case they read. Those five questionnaires were removed from the sample. A negative evaluation outcome received a 0 value and a positive outcome resulted in a 1 value for the created nominal variable.
5.3 Dependent variables
5.3.1 Factor analysis
The Factor analysis shows that the first three questions shown in the questionnaire should be seen together as one component (Evaluation Attitude) and the second three questions as one as well (Fairness Combined).
Table 5: Factor Analysis of Fairness Combined and Attitude towards Fairness
Item (n = 53 for each) Factor Loadings
Fairness Combined Evaluation Attitude
1. I find it right that employees receive a work evaluation
,874
2. I find it right that employees performances are monitored
,801
3. I find it fair that employees are evaluated based on how they perform
,228 ,749
4. I find the system that the firm uses to evaluate Marc fair
,896
5. I find the outcome of the evaluation of Marc based on his performance fair
,903
6. I find the performance evaluation as a whole, as described in the situation before, fair
,909 ,108
Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization
5.3.2 Reliability analysis
As a reliability analysis of the scale of the data, Cronbach’s Alpha test is executed. As shown in table 6 below, the Cronbach’s Alpha of both scales is shown. The Cronbach Alpha is 0,730 for the Evaluation Attitude factor. For Fairness Combined the Cronbach Alpha is 0,892. Both of these are above the Cronbach Alpha’s desired threshold of 0,7.
To analyse if all the items used in the questionnaire are useful, the Cronbach’s Alpha if item deleted is included. For the Evaluation Attitude scale only the third item might be an option to be excluded, as deleting one of the other two items results in a Cronbach’s Alpha below the threshold of 0,7. But as excluding item three will lead to a slightly lower Cronbach’s Alpha for Evaluation Attitude, this is not executed. As far as the Fairness Combined scale goes, deleting one of the three items will still result in a good Cronbach Alpha. However, as was the result for Evaluation Attitude, there is no improvement of the Cronbach’s Alpha compared to keeping all three items in the scale.
22
Table 6: Reliability analysis of Evaluation Attitude and Fairness Combined
Cronbach's Alpha if item deleted Item (n = 53 for each) Evaluation Attitude Fairness Combined
Cronbach's Alpha
0,730 0,892
1. I find it right that employees receive a work evaluation
0,546 2. I find it right that employees performances
are monitored
0,684 3. I find it fair that employees are evaluated
based on how they perform
0,708 4. I find the system that the firm uses to
evaluate Marc fair
0,847 5. I find the outcome of the evaluation of Marc
based on his performance fair
0,843 6. I find the performance evaluation as a whole,
as described in the situation before, fair
0,848
5.3.3 Normality assumption
For an Independent samples T-test, which is used for analysis of the data, it is assumed that the data is normally distributed. This assumption will be tested in this paragraph.
Skewness is used in order to find out if the data is distributed in a symmetrical way. The skewness of Evaluation Attitude is -1,68 for negative evaluation outcome and -0,23 for positive evaluation outcome. For Fairness Combined the numbers are -0,51 and -0,93. This means that there is skewness to the left, as the numbers are negative.
To find out if the skewness of the data has a significant influence, the Z-score of the skewness is calculated. A normal distribution for skewness is assumed when the Z-score is between -2 and +2 (George & Mallery, 2010). The Z-score with the highest value away from 0 is the negative evaluation outcome of Evaluation Attitude (Z-score = -3,62. This indicates that the data that is used has an unacceptable negative skewness. This is also the case for Fairness Combined with a positive outcome (Z-score = -2,11).
Besides skewness, the how tailed a distribution is, is analysed with the kurtosis. As is the case with skewness, a Z-score between -2 and +2 indicates that the data is normally distributed (George & Mallery, 2010). The highest Z-score for kurtosis is for the variable Evaluation
Attitude with a negative evaluation outcome, which is 5,10. This indicates that the data is not distributed normally. The other three Z-scores are well within the acceptable range that indicates a normal distribution of the data.
Table 7: Skewness and Kurtosis
Evaluation Attitude Fairness Combined Variable Negative (n=25) Positive (n=28) Negative (n=25) Positive (n=28) Skewness -1,68 -0,23 -0,51 -0,93 Std. Err. Skewness 0,46 0,44 0,46 0,44
Z-Score Skewness (a) -3,62 -0,53 -1,1 -2,11
Kurtosis 4,60 -0,16 -0,66 0,18
Std. Err. Kurtosis 0,90 0,86 0,90 0,86
Z-Score Kurtosis (b) 5,10 -0,18 -0,73 0,21
(a). Z-score Skewness = skewness/standard error skewness (b). Z-score Kurtosis = kurtosis/standard error kurtosis
Apart from the numbers, above explained effects can also be seen when examining the data in a visual way. With the histograms of the dependent variables, it shows that the data is quite skewed for the Attitude Evaluation with negative evaluation outcome and a little bit for Fairness Combined with a positive evaluation outcome.
24
Figure 1:
Histograms and P-Plots (Negative evaluation outcome)
Attitude Evaluation Fairness Combined
Histograms and P-Plots (Positive evaluation outcome)
For further normality testing, I executed the Kolmogorov-Smirnov and Shapiro-Wilk tests. With a sample size above 50, I will use the Kolmogorov-Smirnov test, as the Shapiro-Wilk test is seen as the more accurate test for sample sizes smaller than 50 (Field, 2009). The Kolmogorov-Smirnov test shows that there are no normality problems with the Evaluation Attitude with the positive outcome condition (p > 0,05). The same applies to the Fairness Combined variable with the negative outcome condition. The other two, Evaluation Attitude with negative evaluation outcome and Fairness Combined with positive evaluation outcome, show significant problems with the normality assumption. By using nonparametric tests for analysis, this problem can be mitigated, as those tests do not assume that the data is distributed normally.
Table 8: Normality tests of Evaluation Attitude and Fairness Combined
Kolmogorov-Smirnova Shapiro-Wilk
Negative (n=25) Positive (n=28) Negative (n=25) Positive (n=28) Variable Statistic Sig. Statistic Sig. Statistic Sig. Statistic Sig.
EvaluationAttitude 0,18 0,04 0,10 0,20* 0,87 0,00 0,96 0,30
FairnessCombined 0,13 ,20* 0,18 0,02 0,94 0,15 0,91 0,02
* This is a lower bound of the true significance
a
Lilliefors Significance Correction
Most of the data that I collected seems to be distributed normally. However, above administered normality tests show that not all the data is normally distributed. Because of this, I will be using parametric tests for my analysis to mitigate this issue, as non-parametric tests do not need the assumption of a normal distribution of the data.
5.3.4 Descriptive statistics
In the below shown table, an overview of the descriptive statistics of the created variables Evaluation Attitude and Fairness Combined. The mean of the Evaluation Attitude is 5,64. This indicates that the respondents have a positive attitude towards evaluating work performances, as the data was collected with a seven-point Likert scale. The same assumption can be made for the Fairness Combined variable.
Table 9: Descriptive statistics of dependent variables
N Minimum Maximum Mean Std. Deviation
EvaluationAttitude 53 2,00 7,00 5,64 0,94
26 The respondents got shown one of two cases. This resulted in two groups to compare. Descriptive statistics of these two groups are shown in table 9. The first group, which contains 25 respondents (47%), read a case with a negative evaluation outcome for Marc. The second group (28 respondents, 53%) got presented a case with the situation where Marc received a positive evaluation outcome. The positive evaluation outcome group showed a slightly higher mean for both dependent variables. The negative evaluation outcome group shows a mean of 5,56 for Evaluation Attitude and a 4,13 mean for Fairness Combined. The group with the positive evaluation outcome shows means of 5,71 and 4,46 for Evaluation Attitude and Fairness Combined.
To be able to visually analyse the outcomes, boxplots are shown below. The boxplot for Evaluation Attitude shows that there is not much difference between both groups of respondents. However, the negative group has two outliers, one of which has a really negative attitude towards performance evaluations in general. The boxplot of the Fairness Combined indicates that respondents that received the case with the positive performance evaluation outcome have a higher perceived fairness of the evaluation. No outliers influence the results in this situation, which means that there are no extreme values having a big influence on the analysis.
There seems to be no clear differences between the two groups of the experiment after reviewing the descriptive statistics and visualizing the outcomes for the Evaluation Attitude. The difference between both of the groups on Fairness Combined is visible.
5.3.5 Correlation
To test if there is an unexpected between variables of the research, a correlation test is executed. As discussed before, the data is skewed and there were two outliers found in the boxplot of Evaluation Attitude from the negative performance evaluation group. Therefore the non-parametric Spearman correlation test will be used. The Pearson correlation test is also shown.
Table 10: Correlation table with Spearman and Pearson correlations
Age Gender Education WExp EvExp EA FC
Age -0,031 (0,826) -0,113 (0,420) 0,846** (0,000) 0,517** (0,000) 0,249 (0,072) 0,134 (0,339) Gender 0,049 (0,730) 0,070 (0,617) -0,098 (0,484) -0,191 (0,171) -0,453** (0,001) -0,154 (0,271) Education 0,015 (0,917) 0,081 (0,566) 0,020 (0,884) -0,072 (0,611) -0,008 (0,956) 0,101 (0,470) WorkExperience 0,917** (0,000) -0,092 (0,512) 0,042 (0,765) 0,619** (0,000) 0,287* (0,037) 0,071 (0,615) EvaluationExp 0,567** (0,000) -0,179 (0,200) -0,068 (0,627) 0,614** (0,000) 0,363** (0,008) -0,063 (0,654) EvaluationAttitude 0,208 (0,135) -0,380** (0,005) 0,052 (0,712) 0,319* (0,002) 0,299** (0,030) 0,237 (0,087) Fairnesscombined 0,113 (0,420) -0,163 (0,244) 0,119 (0,397) 0,094 (0,503) -0,060 (0,667) 0,192 (0,169)
Above the diagonal, the Spearman correlations are shown, whilst the Pearson correlations are presented below the diagonal.
** Correlation is significant at the 0,01 level (2-tailed) * Correlation is significant at the 0,05 level (2-tailed)
Correlations between age, work experience and evaluation experience makes sense, as older people generally have worked for a longer period of time and therefore also have had more experience in being evaluated based on work performance. The other correlations that are found are all connected to Evaluation Attitude. Work experience and evaluation experience are positively correlated, so people who have worked longer have a more positive mindset towards employee evaluations based on work performance. Gender is also correlated to Evaluation Attitude. The male respondents have a more positive stance on Evaluation Attitude than the female respondents. However, as there is no significant difference between
28 male and female respondents in both groups (paragraph 5.1), this will not have a big influence on the analysis.
5.3.6 Hypothesis test
The analysis of the statistical tests to find out whether or not I found empirical evidence of my hypothesis will be discussed in this paragraph.
Hypothesis: Employees with a less favourable performance evaluation outcome will have a lower perceived fairness than employees with a more favourable performance evaluation outcome.
The hypothesis is that the fairness perception of respondents is higher in the group who got the case with the positive evaluation outcome. An independent samples T-test is used to test this. The dependent variable in this test is Fairness Combined. Levene’s test for equality of variance is not significant, so homogeneity of variance can be assumed. The standard deviations of the negative and positive group are 1,37 and 1,15. As there is little difference between both, the assumption of homogeneity of variance stands.
Respondents with a positive evaluation outcome had a slightly higher fairness perception (Mean = 4,46; Std. Err = 0,22) than the people who received the case with the negative evaluation outcome (Mean = 4,13; Std. Err. = 0,27). This test has a t-statistic of -0,956. This difference was not significant, as the p-value is 0,344. As normal distribution issues were found earlier, a bootstrapped independent samples t-test (1000 samples) was executed. This resulted in a p-value of 0,347. Field (2009) clarifies that it is best to see if the outcome is significant by using the 95% confidence interval of the bootstrapped t-test is outside. If zero does not fall within the confidence interval, the results are statistically significant. In this case the confidence interval is [-1,00;0,36], which means that there is no significant difference between the fairness perceptions of both respondent groups.
Another non-parametric test was used to test the hypothesis. The Mann-Whitney U test shows that there is no significant difference (p = 370) between the two experimental groups.
The above applied tests conclude that there is no significant difference between the perceived fairness of employees whether they receive a more favourable performance evaluation or a less favourable outcome. The bootstrapped t-test and non-parametric test show that this finding is robust, as both tests did not show values anywhere near significance.
As an additional analysis, The three questions that were used to create the Fairness Combined dependent variable are checked. With these tests of procedural fairness, outcome fairness and total fairness it is tested if one of the questions does have a significant influence, but that this was mitigated by the other questions used. The means of the three individual questions are really close to the mean of the dependent variable Fairness Combined. The biggest mean differences (Δ Mean = -0,23 for negative and Δ Mean = -0,33 for positive) are between Outcome Fairness and Fairness Combined. But neither for Outcome Fairness, nor for the other two individual variables are the differences between both respondent groups significant, as the p-value that comes closest to significance is 0,229.
Figure X shows a scatterplot of both respondent groups. The green dots represent the respondents who filled out the case with the positive evaluation outcome and the red dots represent the negative evaluation outcome respondents. Both colour dots are mixed together and there is not really a big difference in the location of the spread. Both groups are high on the y-axis, which shows that almost all respondents have a positive attitude towards evaluations based on work performance. Both groups are quite spread out on the x-axis. This shows that the positive evaluation outcome group do not perceive the evaluation outcome as fairer than the negative evaluation outcome group.
30
Table 11: Descriptive statistics tests of created variables of both respondent groups Evaluation Outcome
Evaluation Attitude Negative (n = 25) Positive (n = 28)
Mean Δ 0,15 5,56 5,71 Std. Deviation Δ -0,21 1,05 0,84 Std. Error Mean Δ -0,05 0,21 0,16 Levene's test 0,758 T statistic -0,594 Degrees of freedom 51 Significance (2-tailed) 0,555
Bootstrapped sig. (1000 samples) 0,566
Mann-Whitney U test 0,822 Lower: 95% Confidence -0,676 Upper: 95% Confidence 0,367 Fairness Combined Mean Δ 0,33 4,13 4,46 Std. Deviation Δ -0,22 1,37 1,15 Std. Error Mean Δ -0,05 0,27 0,22 Levene's test 0,283 T statistic -0,956 Degrees of freedom 51 Significance (2-tailed) 0,344
Bootstrapped sig. (1000 samples) 0,347
Mann-Whitney U test 0,370
Lower: 95% Confidence -1,026
Table 12: Tests and descriptive statistics of three individual questions of respondent groups
Evaluation Outcome
ProceduralFairness Negative (n = 25) Positive (n = 28)
Mean Δ 0,42 4,08 4,50 Std. Deviation Δ -0,01 1,44 1,43 Std. Error Mean Δ -0,02 0,29 0,27 Levene's test 0,403 T statistic -1,065 Degrees of freedom 51 Significance (2-tailed) 0,292 Mann-Whitney U test 0,326 Lower: 95% Confidence -1,212 Upper: 95% Confidence 0,372 OutcomeFairness Mean Δ 0,43 4,36 4,79 Std. Deviation Δ -0,21 1,38 1,17 Std. Error Mean Δ -0,06 0,28 0,22 Levene's test 0,279 T statistic -1,217 Degrees of freedom 51 Significance (2-tailed) 0,229 Mann-Whitney U test 0,265 Lower: 95% Confidence -1,128 Upper: 95% Confidence 0,277 TotalFairness Mean Δ 0,15 3,96 4,11 Std. Deviation Δ -0,33 1,62 1,29 Std. Error Mean Δ -0,08 0,32 0,24 Levene's test 0,137 T statistic -0,368 Degrees of freedom 51 Significance (2-tailed) 0,714 Mann-Whitney U test 0,920 Lower: 95% Confidence -0,950 Upper: 95% Confidence 0,655
32
6 Discussion and conclusion
The hypothesis of my research was that if an employee received a more favourable performance evaluation, the employee would perceive the evaluation process as fairer than an employee who received a less favourable performance evaluation. I found no significant evidence in the data that a more favourable evaluation outcome leads to a higher perceived fairness. The same goes for less favourable evaluation outcome and a lower perceived fairness. This is not consistent with the prediction made in the hypothesis chapter.
A more specific analysis was also performed by individually testing the outcomes of the three questions that together formed the Fairness Combined variable. The test results of these questions also show no significant evidence that a less favourable outcome leads to a lower perceived fairness.
Besides the hypothesis, an analysis on the respondents’ attitude towards performance evaluations was conducted. By doing this, I got an image of the respondents’ perceived fairness of being evaluated based on work performance, before they received one of both versions of the manipulated case. This analysis shows that almost all respondents find being evaluated based on work performance a solid concept of performance evaluation. Between both respondent groups, there were no significant differences, so the respondents’ attitudes towards performance evaluations has had no influence on the hypothesis that has been tested. The internal validity of the research can be a problem as it might be too obvious which part and how I used a manipulation to be able to compare both respondent groups. However, using an experiment with a less obvious manipulation is not an option as it will not be clear to respondents what has been altered. This validity concern is a limitation that is inherent to experimental research.
Aside from the internal validity, external validity is also a concern. This research also focussed on one small part of the broader concept of performance evaluations, namely bonus payments based on accounting performance measures. Respondents might not have come across a performance evaluation concept like this before and therefore have a different view of this than respondents that have experience with a performance evaluation structure like this.
Aside from the validity concerns, there are some limitations to the generalization of the results of this research. The respondents that filled out the questionnaire are people from my
social circle. Even though this is quite a mixed group of people, the small sample size should be kept in mind. Another limitation is the culture of the Dutch people. In The Netherlands, people are really direct in their ways of expressing themselves on almost all subjects. When conducting similar research in other countries, this cultural difference should be kept in mind. A suggestion for further research is that one could test whether outcome favourability of a performance evaluation does have an influence on other types of (non-financial) performance evaluation measures. One would expect that outcome favourability of subjective performance evaluation measures would have a significant influence on the perceived fairness of said performance evaluation, when there is less performance evaluation information available to rely on.
34
References
Bergstresser, D., & Philippon, T. (2006). CEO incentives and earnings management. Journal of Financial Economics, 80(3), 511-529. doi:10.1016/j.jfineco.2004.10.011
Bies, R.J (1985). Individual reactions to corporate recruiting encounters: The importance of fairness. Unpublished manuscript cited in Bies and Moag (1986).
Bies, R. J., and Moag, J. S. (1986). Interactional justice: Communicating criteria of fairness. In Lewicki, J., Sheppard, B. H., and Bazerman, M. H. (eds.), Research on Negotiations in Organizations, JAI, Greenwich, CT, pp. 43-55.
Brockner, J., & Wiesenfeld, B. M. (1996). An integrative framework for explaining reactions to decisions: Interactive effects of outcomes and procedures. Psychological Bulletin, 120(2), 189-208. doi:10.1037/0033-2909.120.2.189
Burns, N., & Kedia, S. (2006). The impact of performance- based compensation on misreporting. Journal of Financial Economics, 79(1), 35-67. doi:10.1016/j.jfineco.2004.12.003
Chenhall, R. H. (2003). Management control systems design within its organizational context: Findings from contingency-based research and directions for the future. Accounting, Organizations and Society, 28(2), 127-168. doi:10.1016/S0361-3682(01)00027-7
Cropanzano, R., & Mitchell, M. S. (2005). Social exchange theory: An interdisciplinary review. Journal of Management, 31(6), 874-900. doi:10.1177/0149206305279602
Efendi, J., Srivastava, A., & Swanson, E. P. (2007). Why do corporate managers misstate financial statements? the role of option compensation and other factors. Journal of Financial Economics, 85(3), 667-708. doi:10.1016/j.jfineco.2006.05.009
Field, A. (2009). Discovering statistics using SPSS Sage publications.
Folger, R., Konovsky, M. A., and Cropanzano, R. (1992). A due process metaphor for performance appraisal. In Staw, B. M., and Cummings, L. L. (eds.), Research in Organizational Behavior, Vol. 14, JAI, Greenwich, CT, pp. 129-177.
Folger, R., & Cropanzano, R. (1998). Organizational justice and human resource management. Beverly Hills, CA: Sage
Francis‐Gladney, L., Magner, N. R., & Welker, R. B. (2010). Does outcome favorability affect procedural fairness as a result of self‐ serving attributions? Journal of Applied Social Psychology, 40(1), 182-194. doi:10.1111/j.1559-1816.2009.00569.x
Gaver, J. J., Gaver, K. M., & Austin, J. R. (1995). Additional evidence on bonus plans and income management. Journal of Accounting and Economics, 19(1), 3-28. doi:10.1016/0165-4101(94)00358-C
George, D., & Mallery, P. (2010). SPSS for windows step by step. A simple guide and reference, boston, MA. USA: Ed.
Goldman, E., & Slezak, S. L. (2006). An equilibrium model of incentive contracts in the presence of information manipulation. Journal of Financial Economics, 80(3), 603-626. doi:10.1016/j.jfineco.2005.05.007
Greenberg, J. (1993). The social side of fairness: Interpersonal and informational classes of organisational justice. In R. Cropanzano (Ed.), Justice in the workplace: Approaching fairness in human resource management (pp. 77-103). Hillsdale, NJ: Lawrence Erlbaum Associates.
Hartmann, F. G. H. (2000). The appropriateness of RAPM: Toward the further development of theory. Accounting, Organizations and Society, 25, 451-482.
Healy, P. M. (1985). The effect of bonus schemes on accounting decisions. Journal of Accounting and Economics, 7(1), 85-107. doi:10.1016/0165-4101(85)90029-1
Holthausen, R. W., Larcker, D. F., & Sloan, R. G. (1995). Annual bonus schemes and the manipulation of earnings. Journal of Accounting and Economics, 19(1), 29-74. doi:10.1016/0165-4101(94)00376-G
Hopwood, A. G. (1972). An empirical study of the role of accounting data in performance evaluation. Journal of Accounting Research, 10, 156-182. doi:10.2307/2489870
Latham, G. P., Almost, J., Mann, S., & Moore, C. (2005). New developments in performance management. Organizational Dynamics, 34(1), 77-87. doi:10.1016/j.orgdyn.2004.11.001 Lau, C. M., Low, L. C., & Eggleton, I. R. C. (1995). The impact of reliance on accounting performance measures on job- related tension and managerial performance: Additional evidence. Accounting, Organizations and Society, 20(5), 359-381. doi:10.1016/0361-3682(94)00031-P
36 Lau, C. M., & Sholihin, M. (2005). Financial and nonfinancial performance measures: How do they affect job satisfaction?
doi:http://dx.doi.org.proxy.uba.uva.nl:2048/10.1016/j.bar.2005.06.002
Lau, C. M., Wong, K. M., & Eggleton, I. R. C. (2008). Fairness of performance evaluation procedures and job satisfaction: The role of outcome‐ based and non‐ outcome‐ based effects. Accounting and Business Research, 38(2), 121-135. doi:10.1080/00014788.2008.9663325 Lind, E. A., & Tyler, T. R. (1988). The social psychology of procedural justice. New York: Plenum Press.
Lind, E. A., & Bos, K. (2002). When fairness works : Toward a general theory of uncertainty management. Research in Organizational Behavior, 24, 181-3085.
Locke, E. A., & Latham, G. P. (2002). Building a practically useful theory of goal setting and task motivation. American Psychologist, 57(9), 705-717. doi:10.1037/0003-066X.57.9.705 Merchant, K. A. (1990). The effects of financial controls on data manipulation and management myopia. Accounting, Organizations and Society, 15(4), 297-313. doi:10.1016/0361-3682(90)90021-L
Rawls, J. (1971). A theory of justice. Cambridge, Mass: Belknap Press of Harvard University Press.
Skitka, L., Winquist, J., & Hutchinson, S. (2003). Are outcome fairness and outcome favorability distinguishable psychological constructs? A meta-analytic review. Social Justice Research, 16(4), 309-341. doi:10.1023/A:1026336131206
Thibaut, J., & Walker, L. (1975). Procedural justice: A psychological analysis. Hillsdale, NJ: Erlbaum.
Van, D. B., Lind, E. A., Vermunt, R., & Wilke, H. A. M. (1997). How do I judge my outcome when I do not know the outcome of others? the psychology of the fair process effect. Journal of Personality and Social Psychology, 72(5), 1034-1046. doi:10.1037/0022-3514.72.5.1034
Appendix 1: case plus questions in English
Introduction:
For my thesis I am researching employee evaluations. Therefore I want to ask you to fill out this questionnaire. The questionnaire consists of two parts. First there is a case with corresponding questions. After that there will be some short general questions. All filled out data will be handled with confidentially and answers will not be traceable to an individual.
Thank you for your cooperation.
---
Case
Marc works as a sales professional at the sales office of a trading company. The company sells its products only in the Netherlands. The company is operating in the machine parts industry. Within this company there is a performance evaluation system in place. The system is being used to evaluate employees based on their performance every year. As an employee, Marc has a lot of control and influence on the sales he makes during the year, so hard work will pay off. Below is the firm´s performance evaluation system explained. The system measures an employee’s performance based on the revenue generated by the employee. The evaluation system consists of 5 levels of performance, A through E. The outcome of the evaluation is used to decide what actions need to be taken, based on the employee’s sales performance.
Level Revenue Consequense of evaluation
A > € 450.000 Performance bonus of 10% of salary B € 400.000 - € 450.000 Performance bonus of 5% of salary C € 350.000 – € 400.000 No performance bonus
D € 300.000 – € 350.000 No performance bonus, employee has to follow improvement plan.
E < € 300.000 No performance bonus, employee has to follow improvement plan. After a second straight E evaluation the employee will be fired.
38 --- Please fill out the questions below based on your personal ideas
1. I find it right that employees receive a work evaluation 1 2 3 4 5 6 7
2. I find it right that employees performances are monitored 1 2 3 4 5 6 7
3. I find it fair that employees are evaluated based on how they perform 1 2 3 4 5 6 7
--- 1 2 3 4 5 6 7 Strongly disagree Disagree Somewhat disagree Neutral Somewhat agree Agree Strongly agree
Evaluation outcome 1:
Last year Marc generated € 325.000 in revenue. With this revenue his evaluation has a D value. Because of this rating, Marc is not qualified for a performance bonus. He also has to follow an improvement plan to improve his performance in the coming year.
---
4. Manipulation check: How favourable was your performance evaluation?
Unfavourable 1 2 3 4 5 Favourable
Imagine if you were in the situation of Marc. I would like you to indicate your opinion on the following statements.
5. I find the system that the firm uses to evaluate Marc fair 1 2 3 4 5 6 7
6. I find the outcome of the evaluation of Marc based on his performance fair 1 2 3 4 5 6
7
7. I find the performance evaluation as a whole, as described in the situation before, fair
1 2 3 4 5 6 7
--- Finally a few questions about your personal situation:
6. How old are you? ______ year
7. Are you male or female? O Female O Male
8. What is highest finished education? O University (Bachelor or master) O Professional Education (HBO) O Intermediate Education (MBO) O High school
O Other, namely:______________
9. How many years of work experience do you have (full time and/or part time)?
________ year
10. How many years of work experience do you have being evaluated on your performance in a work environment?
40
Evaluation outcome 2:
This year Marc generated € 425.000 in revenue. With this revenue his evaluation has a B value. Because of this rating, Marc will receive a performance bonus of 5% of his salary. ---
4. Manipulation check: How favourable was your performance evaluation?
Unfavourable 1 2 3 4 5 Favourable
Imagine if you were in the situation of Marc. I would like you to indicate your opinion on the following statements.
5. I find the system that the firm uses to evaluate Marc fair 1 2 3 4 5 6 7
6. I find the outcome of the evaluation of Marc based on his performance fair 1 2 3 4 5 6
7
7. I find the performance evaluation as a whole, as described in the situation before, fair
1 2 3 4 5 6 7
--- Finally a few questions about your personal situation:
6. How old are you? ______ year
7. Are you male or female? O Female O Male
8. What is highest finished education? O University (Bachelor or master) O Professional Education (HBO) O Intermediate Education (MBO) O High school
O Other, namely:______________
9. How many years of work experience do you have (full time and/or part time)?
________ year
10. How many years of work experience do you have being evaluated on your performance in a work environment?
Appendix 2: case plus questions in Dutch
Introductie:
Voor mijn masterscriptie doe ik een onderzoek naar medewerker evaluaties (werkbeoordelingen). Ik wil je daarom vragen om deze enquete in te vullen. De enquete bestaat uit twee delen. Eerst wordt er een casus beschreven, die gevolgd wordt door bijbehorende vragen. Daarna volgen nog een paar korte algemene vragen.
Alle ingevulde gegevens worden vertrouwelijk verwerkt en antwoorden zullen niet te herleiden zijn tot een individu.
Alvast bedankt voor je medewerking.
---Casus
Marc werkt als verkoop professional op de verkoopafdeling bij een handelsonderneming. Dit bedrijf verkoopt allerlei verschillende machineonderdelen binnen Nederland. Binnen dit bedrijf is een prestatie evaluatie systeem in gebruik. Het systeem wordt gebruikt om medewerkers jaarlijks te beoordelen op basis van hun prestaties. Als medewerker heeft Marc veel invloed op de omzet die hij genereert gedurende het jaar, dus hard werken loont. Hieronder wordt het prestatie evaluatie systeem van het bedrijf uitgelegd. Het systeem meet prestaties op basis van de omzet van de desbetreffende medewerker. De beoordeling wordt gebaseerd op basis van 5 niveaus, A tot en met E. De uitkomst van deze beoordeling wordt gebruikt om te beslissen wat voor acties er ondernomen worden, gebaseerd op de prestatie van de medewerker.
Niveau Omzet Gevolg van evaluatie
A > € 450.000 Prestatiebonus van 10% van salaris B € 400.000 - € 450.000 Prestatiebonus van 5% van salaris C € 350.000 – € 400.000 Geen prestatiebonus
D € 300.000 – € 350.000 Geen prestatiebonus, werknemer moet verbetertraject volgen
E < € 300.000 Geen prestatiebonus, werknemer moet verbetertraject volgen. Bij 2e opeenvolgende E beoordeling volgt ontslag
