“How do performance measurement and reward systems affect goal congruence between employees and the organization they work for?”

(1)

Research Project for Master Thesis

“How do performance measurement and reward systems

affect goal congruence between employees and the

organization they work for?”

Master’s Thesis BA O&MC EBM859A20

2015-2016.2

Leonie Vorenkamp S2659417

Supervisor: Wesley Kaufmann University of Groningen

(2)

Executive summary

Despite advances in understanding enabling and coercive management in recent years, there are no studies that have explicitly tested the dependence of goal congruence between employees and the organization they work for on a performance measurement system and a reward system. In this study, I research this gap and argue that goal congruence is dependent on the type of performance measurement system and reward system. Moreover, I argue that if a coercive performance measurement system is used, goal congruence can be improved with a high reward system. Using an experiment (n=102), I show how variations in goal congruence are affected by the performance measurement system and reward system. The results indicate that goal congruence is higher in response to an enabling performance measurement system and a high reward system. Furthermore, there is a significant interaction effect of the performance measurement system and reward system on goal congruence. The negative effects on goal congruence of a coercive performance measurement system can be improved by a high reward system. As a result, it is important to recognize that, although goal

congruence stresses the role of a performance measurement system, reward systems also matter.

(3)

1. Introduction

Management control research in recent years has developed the idea of Management Control Systems (MCSs) as a package which consist of five types of controls that interact (Malmi and Brown, 2008). These five types of control are planning, cybernetic, reward, compensation, administrative and cultural controls (Malmi and Brown, 2008). There have been regular calls to study the MCS as a package, but despite this little explicit empirical research on the topic has been done (Chenhall, 2003) (Malmi and Brown, 2008). Therefore, this research project will study this phenomenon. The main focus of a MCS is how to design a MCS that produces the desired organizational outcomes described in organizational goals and objectives (Malmi and Brown, 2008). However, despite the fact that all organizations have a MCSs, predominant notions of management control research state that MCSs have a number of different definitions and descriptions which makes it difficult to define what is actually meant by a MCS. For example, Merchant and Van der Stede (2012) describe management control as dealing with employees’ behaviour, while Malmi and Brown (2008) have a broader view of what constitutes a MCS and combine management control with strategic control. MCSs are defined in this research as ‘complete systems that consist of rules, practices, values, systems and other activities organizations use to ensure that the behaviours and decisions of their employees are consistent with the organization’s objectives and strategies (Malmi and Brown, 2008).

According to Cugero and Rosanas (2011), alignment between the interests of employees and the interests of the organization is needed in order for organizations to achieve their organizational goals. In short, Cugero and Rosanas (2011) state that MCSs are designed to achieve the greatest possible goal congruence, such that employees pursue personal goals that are conducive to the organizational goals. Accordingly, goal congruence is defined in this research as the alignment of personal goals of the employees with the goals of the

organization they work for (Cugero and Rosanas, 2011). MCSs consist of, among other elements, performance measurement systems (PMSs). In this research PMSs are described as the evolving formal and informal mechanisms, processes, systems, and networks used by organizations for conveying the key objectives and goals elicited by management, for assisting the strategic process and ongoing management through analysis, planning,

(6)

Stonich (1984), besides performance measurement, reward systems are needed within the MCS to create a mechanism to encourage employees to behave in ways that are truly in the firm’s long-term interest. Prior studies have shown that monetary rewards increase effort and performance through focusing employees’ efforts on the organizational goals (Bonner and Sprinkle, 2002). Consequently, reward systems are defined in this research as procedures, rules, and standards associated with the attachment of monetary rewards, on top of the regular salary, to the achievement of organizational goals in order to motivate and increase the

performance of employees (Malmi and Brown, 2008).

Ahrens and Chapman (2004) claim that PMSs can either be coercive in use - which refers to the stereotypical top-down control approach that emphasizes centralization and preplanning – or enabling in use – which seeks to put employees in a position to deal directly with the inevitable contingencies in their work (Ahrens and Chapman, 2004). Enabling and coercive PMSs have been studied in relation to such diverse topics as design characteristics of enabling formalization (Adler and Borys, 1996)(Ahrens and Chapman, 2004)(Chapman and Kihn, 2009)(Dowling and Leech, 2014)(Jordan and Messner, 2012)(Wouters and Wilderom, 2008), flexibility (Ahrens and Chapman, 2004), efficiency (Ahrens and Chapman, 2004),

information system integration (Chapman and Kihn, 2009) audit support systems (Dowling and Leech, 2014), supply-chain accounting practices (Free, 2007), incomplete performance indicators (Jordan and Messner, 2012) and performance measurement systems (Wouters and Wilderom, 2008). Despite the substantial progress that has been made concerning the

knowledge surrounding enabling and coercive PMSs in recent years (Dowling and Leech, 2014), little attention has been paid to one of the key assumptions in the enabling and coercive PMS literature, namely that the successful design and implementation of enabling and

coercive PMSs depends on the reward system that is associated with the PMS within the MCS (Malmi and Brown, 2008).

The strength of management control lies in the broad scope of the controls in the MCS as a package, rather than the depth of its individual systems (Malmi and Brown, 2008).

(7)

organizational performance (Merchant and Van der Stede, 2012). Even though PMS and reward systems have been researched individually, their combined effect on goal congruence has never been researched. This study aims to contribute to the literature by empirically researching how goal congruence is influenced by PMSs and rewards systems with the following research question:

How do performance measurement and reward systems affect goal congruence between employees and the organization they work for?

To determine how a PMS and reward system affect goal congruence between the employees and the organization that they work for, an experiment will be conducted using vignettes. From a methodological point of view, the experimental study departs from mainstream coercive and enabling research that has predominantly relied on case studies (Adler and Borys, 1996) (Ahrens and Chapman, 2004) (Dowling and Leech, 2014) (Free, 2007) (Jordan and Messner, 2012) (Wouters and Wilderom, 2008). In contrast with case studies, the unit of analysis in this study is the condition rather than the respondent

(8)

2. Literature review and hypotheses

In this chapter, the published literature concerning MCSs, PMSs, reward systems and goal congruence will be discussed. After that, hypotheses will be developed and outlined in a conceptual model. Central to this study are MCSs, which are used to assist organizations in achieving organizational goals (Chenhall, 2003) (Malmi and Brown, 2008). Ouchi (1979) states that the problem of organizations in achieving their organizational goals is the problem of obtaining cooperation among a collection of employees and the organization whom share only partially congruent objectives. Advantages of MCSs are the stimulation of effective and efficient implementation of the pre-determined goals and objectives (Merchant and Van der Stede, 2012) (Pujari, 2016) and the stimulation of employees in improving their performance (Pujari, 2016). On the other hand, management control failures can lead to large financial losses, reputation damage, and possibly even to organizational failure (Merchant and Van der Stede, 2012) (Pujari, 2016). However, designed properly including a PMS and reward system, MCSs influence employees’ behaviour in desirable ways and, consequently, increase the probability of goal congruence and subsequently of the organization achieving its goals (Merchant and Van der Stede, 2012). I will review PMSs (enabling and coercive) and reward systems in relation to goal congruence between organizations and their employees. After that, PMSs and reward systems will be reviewed in relation to each other and their combined effects on goal congruence between organizations and their employees will be discussed.

2.1 Performance measurement system and goal congruence

(9)

measures that do not accurately measure performance, the PMS is worthless and can even be a threat to goal congruence and organizational strategies (Kanji, 2002). Moreover, employee motivation can be negatively influenced because of high standards weighing on the

employees and possible internal competition as a result of a PMS (Atkinson, Waterhouse and Wells,1997). Neely, Gregory and Platts (1995) and Atkinson, Waterhouse and Wells (1997) argue that creating a PMS with effective performance measures is time consuming, but is desirable since it helps employees understand what is expected from them, improves productivity and aligns employee goals with organizational goals, hence goal congruence. Subsequently, this leads to the achievement of organizational goals (Neely, Gregory and Platts,1995) (Atkinson, Waterhouse and Wells,1997).

Adler and Borys (1996) suggest that whether employees regard a PMS as enabling or coercive, depends on the characteristics of the PMS. The characteristics of an enabling PMS are repair, internal and global transparency and flexibility (Wouters and Wilderom, 2008). Firstly, they allow users to repair the formal system in case of a breakdown or problem in the work process (Jordan and Messner, 2012). Repair means that users can mend and improve the work process themselves rather than allowing breakdowns and other non-programmable events to force the work processes to a halt (Wouters and Wilderom, 2008) (Adler and Borys, 1996) (Ahrens and Chapman, 2004) (Chapman and Kihn, 2009). Secondly, enabling systems exhibit internal transparency in the sense that employees are able to see through and

understand the logic of the system (Jordan and Messner, 2012). Furthermore, internal transparency means that users have information on its status (Wouters and Wilderom, 2008) (Ahrens and Borys, 1996) (Ahrens and Chapman, 2004) (Chapman and Kihn, 2009). Thirdly, global transparency is a feature of an enabling system and denotes the extent to which

employees understand the up- and downstream implications of their work (Jordan and Messner, 2012). Fourthly, formal systems enable employees to better manage their work if they allow for some flexibility in terms of how they are used (Jordan and Messner, 2012).

(10)

dissatisfaction, and demotivates employees. On the other hand, it provides needed guidance and clarifies responsibilities, thereby easing role stress and helping individuals be and feel more effective (Adler and Borys, 1996). An enabling PMS, in contrast, designs organizational rules and procedures that reckon with the intelligence of workers and helps committed

employees to do their jobs more effectively, gives them the freedom to innovate and reinforces their commitment (Adler and Borys, 1996) (Ahrens and Chapman, 2004) (Free, 2007).

Balogun and Hailey (2008) argue that there are organizations that can better obtain goal congruence with a coercive PMS than with an enabling PMS. When an organization is in a crisis, under time pressure, or undergoing significant change, the coercive PMS is the most appropriate to use. The reason for this is that coercive PMS makes sure that the organization takes control of the situation, provides critical information and quickly steers employees in the right direction. Moreover, coercive PMS can also be effective when setting boundaries and stopping unwanted behaviour of employees (Balogun and Hailey, 2008). A good example to illustrate this is the military, where a coercive PMS is used in order to minimize uncertainty and risk-taking behaviour, since it is very important for soldiers to have clear responsibilities and to carry out their tasks effectively and quickly (Dugan, 2003). However, in the majority of organizations, Wouters and Wilderom (2008) argue, an enabling PMS makes employees feel more facilitated or motivated by the rules and the systems in place than a coercive PMS. Enabling structures are characterized by flexible, decentralized relationships that encourage task mastery, intergroup participation, explicit communication, and trust (Ahrens and Chapman, 2004). An enabling PMS motivates and enables employees to deliver results and aligns employee activities with organizational strategy, since they are more committed (Ahrens and Chapman, 2004). Accordingly, I come to the following first hypothesis:

H1: A coercive PMS will be negatively related to goal congruence between employees and the organization they work for.

2.2 Reward system and goal congruence

(11)

purpose of a reward system is to control the implementation of strategy, since rewards are attached to the achievement of goals (Malmi and Brown, 2008). Moreover, Edward and Christopher (2006) state that people do not automatically come to work, continue to work, or work hard for an organization. They argue that employees need motivation to share and fulfil organizational goals. Reward systems provide the motivation for the alignment of employees’ natural self-interests with the organization’s objectives, thus for goal congruence (Merchant and Van der Stede, 2012) (Malmi and Brown, 2002) (Edward and Christopher, 2006).

Stonich (1984) argues that an organization’s use of the PMS is stimulated by the reward system within the MCS. Reward systems in combination with a PMS should make clear to each employee how to contribute to the overall strategy (Kanji, 2002). This is because reward systems send powerful signals to an organization’s employees about their performance and should motivate employees to take action that moves the organization towards its

strategic goals (Stonich, 1984) (Malmi and Brown, 2008). However, Milkovich et al (2002) state that when rewards are used continually, these can come to be seen as an entitlement rather than as a motivator. Furthermore, it is easy to get unintended consequences if an organization is not clear enough on the attitudes it is hoping to obtain with the rewards. Nevertheless, Milkovich et al (2002) and Yamoah (2014) state that when employee’s actions towards goals can be objectively measured and clearly linked to the rewards, monetary rewards are an easy and seemingly straightforward way to influence the attitudes of

employees. Organizations must align organizational goals and reward systems to place more value on the contribution of employees to attain goal congruence and consequently the realization of set goals.

(12)

and organizational goals by stimulating employees to achieve organizational goals by providing rewards (Malmi and Brown, 2008) (Merchant and Van der Stede, 2012). Anthony and Govindarajan (2007) summarize that the purpose of rewards is to affect the behaviour of employees in the pursuit of goal congruence. In pursuing goal congruence between employees and the organization they work for, there is a need to motivate the employees by rewarding them for desired actions (Anthony and Govindarajan, 2007). Consequently, my second hypothesis is:

H2: A high reward system will be positively related to goal congruence between employees and the organization they work for.

2.3 Interaction effects performance measurement system and reward system on goal congruence

Organizational goals include the overall objectives of the business, which are established by its management and communicated to its employees. Conflicts may exist between the goals of the organization and the employee, since employees have their own goals like personal financial well-being and job security (Paarlberg and Perry, 2007).

Alignment between the interests of employees and the interests of the organization are needed in order to achieve organizational goals. A MCS which constitutes of a PMS in combination with a reward system, among other elements, encourages employee engagement, productivity, goal alignment and as a result the achievement of goals. Kronberg and Gustafsson (2012) state that there are three major guidelines that should be followed in order for the PMS to create goal congruence. These guidelines indicate that (1) the organizational goals must be clear and must be communicated to the employees, (2) it should be possible to transform and define the organizational goals into individual goals and (3) there must be a connection between the individual goals and the rewards that are provided. Kanji (2002) claims that the PMS and reward system are mechanisms of the organization’s MCS, used to influence employee behaviour. These mechanisms constitute a control system with a focus on

performance evaluation measures to create goal congruence between the organization and its employees (Kanji, 2002).

(13)

Adler and Borys (1996) state that a coercive PMS is created to mandate compliance, enforce adherence to certain standards and punish undesirable behaviour. Adler and Borys (1996) argue that the impact of the type of PMS on the attitudes of employees can be positive or negative concerning goal congruence. In organizations with a coercive PMS, the levels of task identity and autonomy required for intrinsic motivation are low (Adler and Borys, 1996). An enabling type of PMS is one that encourages motivation based on identification and as a result the alignment of individual goals with organizational goals, hence goal congruence (Adler and Borys, 1996). An enabling PMS creates greater understanding among employees regarding the fit of their own personal goals within the greater organizational goals (Ahrens and Chapman, 2004). A coercive PMS fosters dissatisfaction and can be detrimental to the motivation and commitment of employees (Ahrens and Chapman, 2004). Therefore, such organizations must rely on purely extrinsic motivation based on threats and rewards to achieve goal congruence between the organization and its employees (Adler and Borys, 1996).

Merchant (1985) argues that employees can have a negative attitude towards a PMS in combination with a reward system, because they feel fear, pressure and anxiety in achieving organizational goals. As a result, they restrain or counteract the performance measures, thus making it difficult to achieve organizational goals (Merchant, 1985). However, according to Stonich (1976), rewards motivate employees to take action that moves the firm towards its strategic goals. Furthermore, Aamodt (2015) states that the possible negative effects of a coercive PMS can be balanced with the use of a reward system. Each employee responds differently to a coercive PMS, for some coercive power motivates them, but for others it might discourage them. An effective organization can achieve stability concerning goal congruence by balancing between a coercive PMS and reward system (Aamodt, 2015). Consequently, I come to the following third hypothesis:

H3: The combination of a coercive PMS with a high reward system will be positively related to goal congruence between employees and the organization they work for.

(14)

coercive PMS in combination with a high reward system is positively related to goal

congruence. In the following chapter the research design which is used to test the hypotheses will be discussed. This research method will be used to examine how PMS and reward systems affect goal congruence between the organization and its employees.

(15)

3. Methodology

In this chapter I will discuss which method has been used in order to address the hypotheses created in my literature review. Moreover, I will justify why this method was selected as the most appropriate for my research by discussing the advantages and constraints. Furthermore, I will explain how the chosen method has been applied. The developed

hypotheses in the previous chapter involve cause-and effect relationships and according to Jackson (2012), an experiments is an effective technique to evaluate these hypotheses. Additionally, Jackson (2012) states that an experiment is a study in which the researcher manipulates the level of some independent variables and then measures the outcome. The utility of the true experiment is that due to the rigorous controls, exercises, the random assignment of participants to the treatment and the ability to rule out rival explanations, true experiments have a much higher internal validity than other experiment research designs (Liebert and Liebert, 1995). Hence, Liebert and Liebert (1995) argue that findings may be interpreted as having greater integrity. Also, the true experiment is the most suitable of any of the research designs for testing hypotheses involving causal relationships (Williamson, 2002) (Jackson, 2012). Therefore, I have chosen to conduct a true experiment in this study.

3.1 The experiment

The experiment consists of four treatments. Each treatment is a condition containing a scenario about potential goal congruence between the employee and the organization, in which the participant has taken the role of the employee while working for the organization. The scenarios differ on two dimensions: PMS (enabling or coercive) and reward system (low versus high). I expected varying levels of goal congruence for each consecutive category, with the lowest level in Scenario A (coercive performance management system, low reward

(16)

This experiment has been conducted online, even though online studies can make it difficult to develop procedures that ensure internal validity. Possible reasons for this are that instructions may be ignored or read too carelessly and distractions may occur during the course of the experiment (Hoffman and Morgan, 2011). An alternative would be a laboratory experiment, however the external validity is a major weakness of laboratory experiments. When an experiment is conducted in rigidly controlled laboratory conditions, its

generalisability to other populations and settings lacking those controls is uncertain

(Williamson, 2002). Furthermore, a random sample has been used. While a random online sample is not ideal, in that the respondents might not understand the experiment and will not provide relevant input since this is difficult to control, this sample has some advantages. The advantages of a random sample include its ease of use and its accurate representation of the larger population (Van Aken, Berends and Van der Bij, 2012). In order to overcome the weaknesses of an online study with a random sample, the instrument has been administered online to at least 80 people with an attention check, which is a kind of trick question that uses a large block of text in which the respondent is asked to answer in a certain way, to ensure internal validity. The attention check ensures valid responses for my online study, which improves the internal validity since the quality of my research depends on the quality of my data (Van Aken, Berends and Van der Bij, 2012). The attention check can be found in Appendix 7.2.

3.2 The measures

(17)

specified over the five variables which constitute goal congruence (importance, relevance, motivation, commitment, and involvement).

The level of goal congruence is measured through the five questions related to the level of goal congruence concerning quality goals. The reason for quality to be chosen as an

organizational objective is that Eisenhardt (1989) states that rewards represent important mechanisms by which employee behaviours can be aligned with the interests of the organization, but that rewards make employees focus more on quantity than quality. The reason for this is that employees do not normally directly profit from quality in contradiction to profitability, productivity and improved performance like the organization does (Suff, Reilly and Cox, 2007). Respondents have answered the five questions relating to goal congruence based on a 5-point Likert scale and from this the average is taken. With this information the researcher is able to determine the level of goal congruence in relation to different conditions, since it measures whether the employees pursue goals that are conducive to the organizational goals.

3.3 The respondents

Individual respondents have been approached via the social media platform Facebook to participate in the experiment. Furthermore, I have organized and led two group sessions in which respondents have participated in the experiment. The two group sessions have been conducted in a closed room in which all the participants of the group entered one by one to conduct the experiment. The majority of the responses (72,55%) were gathered via the group sessions. There was a possibility of a response bias, since the participants of the group sessions could possibly have been influenced by stimuli created by the group sessions. This could have affected the responses (Van Aken, Berends and Van der Bij, 2012). Therefore, I have controlled for this possibility within my regression analysis.

(18)

gender as control variables and besides this information the experiment was anonymous. The average age of the respondents is 27, with a minimum of 19 and a maximum of 61 and a standard deviation of 7,6. The average gender of the respondents is 1,57 (1= male, 2=female), which indicates that 57% of the respondents is female.

3.4 The analysis

After the experiment has been conducted and all the data has been collected, SPSS will be used to test the hypotheses. Following Pituch and Stevens (2016), the results have first been analysed using descriptive results: This includes the number of times a scenario has occurred within the data set, the mean goal congruence rate per scenario, the minimum and maximum goal congruence rate per scenario and the standard deviation. This is helpful in researching hypotheses 1 and 2. To follow up, an ANOVA has been conducted which will compare the average goal congruence per scenario to test the hypotheses 1 and 2 (Pituch and Stevens, 2016) (Berkman and Reise, 2011). Furthermore, a regression analysis and

MANOVA have been conducted to test all three of the hypotheses.

According to Pituch and Stevens (2016) the regression analysis is a suitable test to predict goal congruence based on the PMS and reward system. Respondents that failed the manipulation check were not filtered out, therefore in the regression analysis there has been controlled for the possibility that the failed manipulation check has an effect on goal

congruence (Berkman and Reise, 2011). Furthermore, the majority of the responses (72,55%) were gathered via the group sessions in the regression analysis. Therefore, I have controlled for the possibility that the results are influenced by the difference of respondents having taken part in the experiment individually or via a group session. The regression analysis has been used to test hypotheses 1 and 2. The MANOVA has been used because there are five measures, since goal congruence in this research consists of the five measures importance, relevance, motivation, commitment and involvement. The MANOVA has been conducted to test whether there is an interaction effect between the independent variables PMS and reward system on goal congruence, which researched hypotheses 3 (Pitch and Stevens, 2016)

(Berkman and Reise, 2011).

(19)

4 Results

This chapter presents the results of the experiment that I have conducted. In this chapter the statistical analysis that has been performed in SPSS is presented, assumptions have been examined and findings have been reported and clearly explained. The results are presented in a way that shows clear support for my hypotheses. During the statistical analysis, gender and age proved not to be significant drivers of goal congruence, their significances are 0,060 and 0,337 respectively.

4.1 Descriptive results

Table 1 provides descriptive statistics for the four experimental treatments. In total, 102 responses following from the experiment have been used in the statistical analysis. The respondents that have failed the attention check have been eliminated to increase the internal validity of the results. Within these 102 experiments, scenario D has occurred 19 times, scenario A occurred 25 times, scenario C occurred 26 times and scenario B occurred 32 times. As expected, I find that average goal congruence is higher for each consecutive scenario (ranging from a value of 2,04 for Scenario A to a value of 4,68 for Scenario D). Furthermore, the minimum of the average goal congruence is higher for almost each consecutive scenario (ranging from a value of 1 for Scenario A to a value of 3 for Scenario C and D). The

maximum varies from a value of 4 for Scenario A to a value of 5 for Scenario B, C and D. The standard deviation decreases with almost each consecutive scenario (ranging from a value of 1,306 in Scenario A to a value of 0,599 in Scenario C). In Scenario D the value of the standard deviation increases again to a value of 0,671.

Table 1 - Descriptive statistics for study variables

Measures N Mean Min Max S.D.

Experimental goal congruence 102 3,56 1 5 1,270

Scenario A 25 2,04 1 4 1,306

Scenario B 32 3,69 2 5 0,738

(20)

Scenario D 19 4,68 3 5 0,671

4.2 ANOVA

The difference in mean values of goal congruence between scenarios is significant, as indicated by the ANOVA results reported in table 2. This result implies that goal congruence is driven both by PMSs and reward systems, providing preliminary support for Hypothesis 1 and 2. However, more robust analysis is required to validate these results, to test hypothesis 3, and to control for the effect of demographic attributes, failed manipulation checks and

individual responses versus group sessions. Therefore, an OLS regression analysis and MANOVA are employed to the data.

Table 2 - ANOVA analysis of experimental goal congruence across scenarios

Average

Sum of squares Df Mean square F Sig. Between groups 95,484 3 31,828 48,490 ,000

Within groups 64,326 98 ,656

Total 159,810 101

4.3 Regression analysis and MANOVA

The regression results are shown in table 3. The dependent variable in my analysis is goal congruence, which is the average from the five variables (importance, relevance,

(21)

better analyse the relative strength of my predictors, I also added a column with standardized coefficients in model 3 on which I focus my discussion.

The earlier findings with regard to the experimental treatment type are confirmed by the regression analysis. All three of the scenarios have a positive coefficient, and are

significant in the full model specification at p<0.05 and p<0.01. This means that the three scenarios with increasing levels of enabling PMS and higher rewards systems (Versions B, C, and D) are positively related to goal congruence when compared to the scenario with coercive PMS and a low reward system (Version A). I also measured the influence of the failure of the two manipulation checks and the influence of the group sessions, however this proved not to be significant. Thus, respondents report higher levels of goal congruence in response to the consecutive scenarios with a high reward system and an enabling PMS, regardless of the influence of the failure of the manipulation checks or the two group sessions.

The MANOVA test (table 4) indicated that reward systems, PMSs and reward system*PMS vary significantly over goal congruence (p < 005). This means that there is a significant interaction between the two independent variables (reward system and PMS) on goal congruence. The most commonly recommended multivariate statistic to use is Wilks’ Lambda (Tinsley and Brown, 2000) and therefore this has been used in this research as well. The values of Wilks’ Lambda indicate that about 39,6% of the variance of goal congruence is accounted for by the differences between reward systems, 58,7% by PMS and only 12,2% by reward system*PMS. Thus, these results suggest that the effects of PMS are larger in size than the effects of the reward system on goal congruence. Overall, I find strong support for

Hypotheses 1, 2, and 3. None of my control variables are significant in the full model specification (table 3, model 2). In the following chapter, the conclusions, I will summarize the work of the research project and conclusions will be formed following the results of the experiment.

Table 3 - OLS Regression predicting goal congruence

Measures Model 1 Model 2 Model 3

(22)

Age ,010 (,019) ,021 (,012) ,128 Significance: ,069 Group session -,121 (,326) -,369 (,202) -,132 Significance: ,071 Manipulation check -,237 (,435) ,035 (,270) ,008 Significance: ,899 Scenario B 1,806 (,211) ,670 Significance: ,000 Scenario C 2,233 (,225) ,778 Significance: ,000 Scenario D 2,838 (,242) ,883 Significance: ,000 Constant 4,406 (1,242) 4,046 (1,042) N 102 102 F Score ,582 24,242 R-Squared ,023 ,644 Adjusted R-Squared -,017 ,617

Table 4 - Multivariate Test

Independent variable Value Percent F Significance Reward system ,604 39,6 % 12,351 ,000

PMS ,413 58,7% 26,686 ,000

Reward system * PMS

(23)

5. Conclusions

In this chapter I will shortly discuss the literature, methodology and results of this

research. After that, I will present the conclusions, point out the limitations of this study, and make relevant indications of areas for further research. Furthermore, the implications that follow from this research for researchers and organizations are presented. This research has been conducted since there was a gap in the literature concerning the effects of the PMS and reward system on goal congruence. Even though PMS and reward systems have been researched individually, their combined effects on goal congruence have never been

researched. Using an experiment with varying PMSs and reward systems, I have set the first step in closing this gap. Consequently, this study has contributed to the literature by

empirically researching how goal congruence is influenced by PMSs and reward systems. This research been done this with the research question ‘How do performance measurement

and reward systems affect goal congruence between employees and the organization they work for?’. Central to this research are MCSs, which are used to assist organizations in

achieving organizational goals. These systems are needed because it is difficult for organizations to obtain cooperation among employees and the organization since their objectives are only partially congruent. A PMS in combination with a reward system, which both are part of a MCS, encourages employee engagement, productivity, goal alignment and the accomplishment of goals.

Hypotheses 1 and 2 are concerned with whether a coercive PMS is negatively related to goal congruence and whether a high reward system is positively related to goal congruence. Hypothesis 3 is concerned with whether a coercive PMS with a high reward system is

(24)

5.1 Limitations and future research

Despite the insights that this study has to offer, it is important to note some limitations of the research at this point. Firstly, the experiment has been administered online. Online studies can make it difficult to develop procedures that ensure internal validity and reliability. The reason for this is that it is difficult to verify the identity of the respondents, instructions may be ignored or read too carelessly and distractions may occur during the course of the experiment. Secondly, and related to the previous point, our experimental setting entails a random online sample. Respondents might not understand the experiment and a result could be that they will not provide relevant input. This is because it is that it is difficult to control an online study, indicating the need for caution when generalizing the findings from this study. However, the attention check within the experiment has improved the internal validity and consequently the generalizability of this study. Thirdly, and related to the previous points, despite the result that the group session has proven not to significantly influence the results concerning goal congruence, the internal validity and reliability could be improved by not having a division in the way respondents participate in the experiment. This reason for this is that this could have led to a response bias, since the respondents of the group sessions could possibly have been affected by stimuli created by the group sessions that could have affected the responses of the respondents. Fourthly, additional research is required to verify to what extent the PMS and findings of this experimental study coincide with ‘real-life’

manifestations of PMSs.

From this limitations, suggestions for future research can be created. Firstly, other research methods can be used while conducting this study. For example, a laboratory research can be done instead of an administering the experiment online and having individual

(25)

influence the results. Therefore, another research method could be a field experiment that has better external generalizability and validity than laboratory studies and case studies.

Furthermore, field experiments can be done in large institutions, which is not possible in laboratory experiments.

Secondly, concerning the respondents, there should be no division between individual respondents and group sessions to increase the internal validity and reliability. The reason for this is that this could influence the study results since this creates different situations in which the respondents participate in the experiment, which could influence the results of the

experiment. Thirdly, the Cronbach Alpha for the five measures that constitute goal

congruence was relatively high (0.97) which was ideal for construct validity. The results in this study indicated a significant interaction effect between PMS and reward system. In order to verify the interaction effect, further research should repeat this study with other measures to test for differences and to find out which measures are particularly suitable to measure goal congruence. Fourthly, and related to the previous point, further research could conduct this research with other vignettes to test for differences and to find out if other vignettes are more suitable to research goal congruence.

5.2 Implications for researchers and organizations

This study also has important implications for researchers. Current enabling and coercive management research says little about its relation with reward systems and goal congruence. This could be researched more extensively in the future in a laboratory setting, case study and/or field experiment with other measures and vignettes. Moreover, future enabling and coercive management research could also integrate other factors of a control system in the research in order to understand the outcomes of goal congruence in

(26)

(27)

6. Reference List

Aamodt, M.G. 2015. Industrial/Organizational Psychology: An applied approach. Boston: Centage Learning.

Adler, P., S., Borys, B. 1996. Two Types of Bureaucracy: Enabling and Coercive.

Administrative Science Quarterly, 41: 61-89.

Ahrens, T., Chapman, C.S. 2004. Accounting for flexibility and efficiency: A field study of management control systems in a restaurant chain. Contemporary Accounting Research, 21:271-301.

Aken, van, J.E., Berends, H., Bij, van der, H. 2012. Problem solving in organizations: a

methodological handbook for business and management students (2nd edition). New York: Cambridge University Press.

Anthony, G., Govindarajan, A.L. 2007. Management control systems. Management

Accounting research, 11: 263-279.

Atkinson, A.A., Waterhouse, J.H., Wells, R.B. 1997. A stakeholder approach to strategic performance measurement. MIT Sloan Management Review, www.sloanreview.mit.edu. Balogun, J., & Hailey, V.H. 2008. Exploring Strategic Change. Harlow: Prentice Hall Berkman, E.T. & Reise, S.P. 2011. A conceptual guide to statistics using SPSS. Thousand Oaks: SAGE Publications

Bonner, S.E., Sprinkle, G.B. 2002. The effects of monetary incentives on effort and task performance: theories, evidence, and a framework for research. Accounting Organizations

and Society, 27: 303–345.

Brewer, G.A., Brewer Jr, G.A. 2011. Parsing Public/Private Differences in Work Motivation and Performance: An Experimental Study. Journal of Public Administration Research and

Theory, 21: 347– 362.

Campbell, D.J. 1995. Motivation theory in industrial and organizational psychology.

Handbook of industrial and organizational psychology. Chicago: Rand McNally.

(28)

Chenhall, R.A. (2003). Management control systems design within its organizational context: findings from contingency-based research and directions for the future. Accounting,

organizations and Society, 28: 127-168.

Cugero, N. Rosanas, J.M. 2011. Fairness, justice, subjectivity, objectivity and goal congruence in management control systems. IESE Business School. www.iese.edu.

Dowling, C., Leech, S.A. 2014. A Big 4 Firm’s Use of Information Technology to Control the Audit Process: How an Audit Support System is Changing Auditor Behaviour. Contemporary

Accounting Research, 31:230-252.

Dugan, M.A. 2003. Coercive Power. Moving Beyond Intractability, 20: 87-95.

Edward, E. L. III, Christopher, G. W. 2006. Winning support for organizational change: Designing employee reward systems that keep on working. Ivey Business Journal Online, www.iveybusinessjournal.com.

Eisenhardt, K.M. 1989. Agency theory: an assessment and review. Academy of Management

Review, 14: 57–74.

Feeney, M.K. 2009. Bureaucracy and public employee behaviour, a case of local government.

Review of public personnel administration, 29:11-326.

Free, C. 2007. Supply-chain accounting practices in the UK retail sector: Enabling or coercing collaboration? Contemporary Accounting Research, 24: 897-933.

Hoffman, M., Morgan, J. 2011. Who’s Naughty? Who’s Nice? Social Preferences in Online Industries. Haas School of Business. www.faculty.haas.berkeley.edu.

Jackson, S.L. 2012. Research Methods and Statistics: A Critical Thinking Approach (4th edition). Wadsworth: Cengage Learning.

Jordan, S., Messner, M. 2012. Enabling control and the problem of incomplete performance indicators. Organizations and Society, 37:544-564.

Kanji, G.K. 2002. Performance measurement system. Total quality management, 13: 715-728.

(29)

Liebert, R. M., Liebert, L. L. 1995. Science and behavior: An introduction to methods of

psychological research. Englewood Cliffs, NJ: Prentice Hall.

Locke, E.A. 2004. Linking goals to monetary incentives. The Academy of Management Executive, 18: 130-133.

Malmi, T., Brown, D.A. 2008. Management control systems as a package – Opportunities, challenges and research directions. Management Accounting Research, 19: 287-300.

Merchant, K.A. 1985. Control in Business Organizations. Marshfield: Pitman.

Merchant, K. A., Van der Stede, W. A. 2012. Management control systems: Performance

measurement, evaluation and incentives (3th edition). Harlow: Pearson Education.

Milkovich, G. T., Newman, J. M., & Milkovich, C. (2002). Compensation (Vol. 8). T. Mirror (Ed.). London: McGraw-Hill.

Mohsan, F., Nawaz, M.M., Khan, M.S., Shaukat, Z., Aslam, N. 2015. Are employee motivation, commitment, and job involvement interrelated: Evidence from banking sector from Pakistan. International Journal of Business and Social Science, 2: 226-233.

Neely, A.D., Gregory, M., Platts, K. 1995. Performance measurement system design.

International Journal of Operations & Production Management, 4: 80 – 116.

Ouchi, W.G. 1979. A conceptual framework for the design of organizational control mechanisms. Management Science, 25: 833-848.

Paarlberg, L.E., Perry, J.L. 2007. Values management: Aligning employee values and organization goals. The American Review of Public Administration, 20: 387-408.

Pituch, K.A. & Stevens, J.P. 2016. Applied multivariate statistics for the social sciences:

analsysis with SAS and IBM’s SPSS. Abingdon: Routhledge.

(30)

Sniderman, P.M., Grob, D.B. 1996. Innovations in Experimental Design in Attitude Surveys.

Annual Review of Sociology, 22: 377–399.

Stonich, P.J. 1984. The performance measurement and reward system: Critical to strategic management. Organizational dynamics, 12: 45-57.

Suff, P., Reilly, P., Cox, A. 2007. Paying for performance. New trends in performance-related pay. Institute for employment studies. www.employement-studies.co.uk.

Tinsley, H. E.A., Brown, S.D. 2000. Handbook of applied multivariate statistics and

mathematical modelling. Cambridge Massachusetts: Academic Press.

Williamson, K. 2002. Research methods for students, academics and professionals,

Information Management and Systems (2nd edition). Wagga Wagga: Centre for Information Studies.

Wouters, M., Wilderom, C. 2008. Developing performance-measurement systems as enabling formalization: A longitudinal field study of a logistics department. Accounting,

Organizations and Society, 33:488-516.

Yamoah, E.E. 2014. Relevance of goal congruence and its implications to managers.

(31)

7. Appendix

7.1 Text of conditions

This manipulation is based on the characteristics of formalization; repair, internal and global transparency and flexibility (Wouters and Wilderom, 2008).

Scenario A – Coercive performance measurement system, low reward system The performance measurement system at the Toy Factory states your duties and output targets at the assembly line, which means there are many written rules describing both the production process and your duties at the assembly line.

When the assembly line breaks down, you are not able to repair it yourself. Furthermore, you do not have access to information on the status of the assembly line.

You do not understand the implications of your work and do not know what happens before or after your role in the assembly line.

The performance measurement system does not enable you to better manage your work because it does not give you any flexibility.

Your performance is measured on a daily basis with respect to output targets.

You will be fired if your performance is below the stated target three times during one month.

Your annual salary at the Toy Factory before taxes is € 24,000. On top of this, you can receive a bonus of up to € 500 based on your performance during the year.

Scenario B – Coercive performance measurement system, high reward system

(32)

When the assembly line breaks down, you are not able to repair it yourself. Furthermore, you do not have access to information on the status of the assembly line.

You do not understand the implications of your work and do not know what happens before or after your role in the assembly line.

The performance measurement system does not enable you to better manage your work because it does not give you any flexibility.

Your performance is measured on a daily basis with respect to output targets.

You will be fired if your performance is below the stated target three times during one month.

Your annual salary at the Toy Factory before taxes is € 24,000. On top of this, you can receive a bonus of up to € 10,000 based on your performance during the year.

Scenario C – Enabling performance measurement system, low reward system

The performance measurement at the Toy Factory tells you how the whole production process works, which means that there are few written rules describing the production process and your duties at the assembly line.

When the assembly line breaks down, you are able to repair it yourself. Furthermore, you have access to information on the status of the assembly line.

You understand the implications of your work and know what happens before or after your role in the assembly line.

The performance measurement system enables you to better manage your work, because it gives you flexibility.

(33)

If your performance is below the stated target three times during one month, your supervisor will schedule a meeting with you to look for ways to improve your performance.

Your annual salary at the Toy Factory before taxes is € 24,000. On top of this, you can receive a bonus of up to € 500 based on your performance during the year.

Scenario D – Enabling performance measurement system, high reward system

The performance measurement at the Toy Factory tells you how the whole production process works, which means that there are few written rules describing the production process and your duties at the assembly line.

When the assembly line breaks down, you are able to repair it yourself. Furthermore, you have access to information on the status of the assembly line.

You understand the implications of your work and know what happens before or after your role in the assembly line.

The performance measurement system enables you to better manage your work, because it gives you flexibility.

Your performance is measured on a monthly basis with respect to output targets.

If your performance is below the stated target three times during one month, your supervisor will schedule a meeting with you to look for ways to improve your performance.

(34)

7.2 Attention check

Organizational culture is a fuzzy concept that is hard to define. To help us understand how people interact in organizations we are interested in how people react to culture. Specifically, we are interested in how much you read instructions; if not, your answers may not tell us much about people in real organizations. To show that you have read these instructions please ignore the question below about organizational culture and check only "None of the above" as your answer.

Please select all that describe the organizational culture that fits your personality best:

o Fun o Exciting o Dreadful o Innovative o Collaborative o Open o Free o Tolerant o Obedient o Oppressive

(35)

7.3 Management Control System Study

1. How important are the quality standards of the Toy Factory to you personally on a scale from 1 to 5 (where 1 means not important at all, and 5 means very important)?

2. How relevant are the quality standards of the Toy Factory to you personally on a scale from 1-5 (where 1 means not relevant at all, and 5 means very relevant)?

3. How motivated are you to safeguard the quality standards of the Toy Factory on a scale from 1-5 (where 1 means not motivated at all, and 5 means very motivated)?

4. How committed are you to pursue the quality standards of the Toy Factory on a scale from 1-5 (where 1 means not committed at all, and 5 means very committed)?

(36)

7.4 Tests of between-Subjects Effects

Source Dependent variable Goal congruence

Df F Significance

Reward system Relevance 3 36,525 ,000

Importance 3 36,305 ,000 Motivation 3 40,833 ,000 Committed 3 45,219 ,000 Involved 3 60,740 ,000 PMS Relevance 3 61,476 ,000 Importance 3 73,803 ,000 Motivation 3 65,608 ,000 Committed 3 63,004 ,000 Involved 3 109,734 ,000

Reward system*PMS Relevance 3 9,040 ,003

Importance 3 7,942 ,006

Motivation 3 5,635 ,020

Committed 3 8,588 ,004

“How do performance measurement and reward systems affect goal congruence between employees and the organization they work for?”

Research Project for Master Thesis