The effect of result
control
on work tension and the role of the
relationship with colleagues in professional service firms.
Name: Rick de Smit
Student number: 11392851 Thesis supervisor: S. P. van Triest
Date: June 19, 2018
Word count: 12,951
MSc Accountancy & Control, specialization Control
Statement of Originality
This document is written by student Rick de Smit 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
This study uses a survey method to investigate the relation between the use of result control and work tension in professional services firms. In particular, this study investigates the role of the relationship with colleagues between result control and work tension. The contribution of this research question to existing knowledge will be that organizations and managers after this research have more insight into the effect of applying result control. Through this research, organizations and managers will better understand the effect of the use of result control on work tension and the importance of a good relationship between employees. A group of 730 professional service employees gave answer on different types of questions that have to do with the extent of the use of result control, the degree of work tension and the relationship they have with their colleagues. To investigate the role of the relationship with other employees between the use of result control and the effect on work tension, the social identity theory is used. Prior studies have shown that when employees identify themselves with others, this could lead in a positive effect by helping other colleagues by providing them with social support or doing extra work (Hopkins, 1997; Tyler, 1999). Haslam (2004) argue that this social support and extra work leads to a better relationship with colleagues and less work tension experienced among the employees. It has been concluded that the relationship with colleagues affect to such an extent that less work tension is experienced by employees when they have a better relationship with colleagues, but work tension still increases when more result controls are used.
Contents
1 Introduction ... 5
2 Theory ... 8
2.1 Professional service firms ... 8
2.2 Management control system ... 9
2.3 Result controls ... 11
2.4 Social Identity Theory ... 12
2.5 Result control and work tension ... 13
2.6 Result control and relationship with colleagues ... 15
2.7 Relationship with colleagues and work tension ... 17
3 Method ... 18 3.1 Respondents ... 18 3.2 Instrument ... 19 3.3 Data-analysis ... 19 3.3.1 Work tension ... 19 3.3.2 Result control ... 20
3.3.3 Relationship with colleagues ... 22
3.3.4 Gender ... 23
3.3.5 Firm Size ... 23
3.3.6 Firm ownership ... 23
3.3.7 Descriptive statistics and correlations matrixes ... 24
4 Results ... 26
5 Discussion ... 30
5.1 Theoretical implications ... 30
5.2 Limitations ... 31
5.3 Suggestions for future research ... 31
5.4 Practical implications ... 32
References ... 33
1 Introduction
This study uses a survey method to investigate the use of result control on work tension in professional service firms. In particular, this study investigates the role of the employee’s relationship with colleagues in relation to the use of result control and work tension. Merchant and van der Stede (2017) stated that result control is about measuring the results of the employees’ effort. The employee is judged on the performance he delivers. It is important to know that result control is only effective when the desired results can be influenced by the employees and the results are measurable (Merchant and van der Stede, 2017). Work tension is related to work related stress, stress can be defined as a person’s individual response to environmental stressors (Vigoda, 2002, p. 574). Work tension or stress, is the degree to which the employees feel that they cannot perform work in the way they believe it should be done. (Dijkhoff, 2016). Where work tension is more related to the inside feelings of a person, the relationship with colleagues is more about how the employees feel related to the other employees. It is about the connectedness they feel with their colleagues.
Previous studies have shown that using (result) controls leads to more work tension (or job-related stress) among employees (Shields et al., 2000, p. 196; Ganster et al., 2011, p. 225). This is usually because when more or higher targets and goals are set for employees, employees experience more stress on their work because they have to perform better than before. In this study, it is expected that by applying more result controls in professional service firms, employees experience a more ‘ingroup-feeling’ because of social comparison, which lead to employees who are more willing to help each other and therefore have a better relationship with other colleagues which leads to less work tension among employees. If this expectation is true, the relationship with other colleagues stated that less work tension is experienced when more result controls are used in professional service firms. To examine this assumption, the following research question in developed: “Does the relationship with colleagues affect the relationship between result control and work tension in professional service firms?”
Professional services are organizations with relatively few transactions, highly customized, process oriented, with relatively long contract time, with most value added in the front-office and where considerable judgment is applied in meeting customer needs (Auzair & Langfield-Smith, 2005, p. 402). Von Nordenflycht (2010, p. 160) distinguished three different characters with regard to professional service firms: (1) knowledge intensity, (2) low capital intensity and (3) professionalized workforce. Employees are essential in professional service firms, given
that on the basis of their knowledge and skills, the organization earns money.
To investigate the role of the relationship with colleagues between result control and work tension in professional service firms, the social identity theory is used. The social identity theory proposes that a person’s sense of who they are depends on the groups to which they belong (Tajfel and Turner, 1986). Tajfel and Turner (1986) describe three processes that can create an ingroup or outgroup mentality, these are: social categorization, social identification and social comparison. In this research, we focus on the aspect of social comparison. This is the phase after a person categorize themselves within a group and identify themselves as being a member of a group. Social comparison is about comparing our group, the ingroup, against another group, an outgroup (David, 2015). It is about the ‘us’ versus ‘them’ mentality. This attitude can lead to a better relationship with colleagues. Prior studies have shown that when employees identify themselves with others, this could lead in a positive effect by helping other colleagues by providing them with social support or doing extra work (Hopkins, 1997; Tyler, 1999). Haslam (2004, p. 170) argue that sharing of labour can help to promote group productivity and with this he suggests that this could lead to better relationship between employees and a better performance for the organization as a whole. On the other side, the social identity theory provides a basis for effective support and less stress is experienced by the employee because employees want to help each other (Haslam, 2004, p. 196). In contrast, the use of result control can also cause more stress to be experienced when the task demands increase over performance capability (Shields et al., 2000, p. 191).
In summary, we could say that by applying more result control, this could lead to both more or less work tension. Normally, we would say that more result control leads to more work tension, according to Shields et al. (2000) and Ganster et al. (2011). But in this research, we expect that when more result controls are used, employees through the social identification and social comparison (‘us’ versus ‘them’) have a better relationship with other colleagues and that therefore employees experience less work tension in their organization. The social identity theory convinced me of the expectation that work tension decreases when employees have a better relationship with their colleagues, despite the fact that more results controls are applied. The contribution of this research question to existing knowledge will be that organizations and managers after this research have more insight into the effect of applying result control. Through this research, organizations and managers will better understand the effect of the use of result control on work tension and understand the role of the employee’s relationship with
their colleagues. This research is a response to Shields et al. (2000) who ask for more research that improve our understanding of the effect of control systems on job related stress (or work tension). In addition, it is also a response to King and Clarkson (2015) who argue that future research has to be done in professional service firms (PSF), especially research that improve our understanding to optimal MCS design to improve performance outcomes (2015, p. 36). But, why would ‘professional service firms’ be worth investigating? From an organizational perspective, it is important that professional service employees do their work good and within the set time. The success of professional service firms depends on their people and their resources. Therefore, organizations make use of result controls in such a way that the employees act in the best interest of the organization. Kuang et al. (2014) found that firms hiring consultants (part of PSE’s) set more challenging performance targets than firms not using a consultant. This will lead to pressure on the work tension of the consultant from both the manager and the client they work for. Managers must therefore realize what the consequences of result control can be. From this perspective, our research question aims to give organizations and managers more insight in the effect of applying result control, especially related to work tension.
This study uses the following hypothetical model (see figure 1) to answer our research question. This model contains the hypotheses and helps to investigate the use of result control on work tension in professional service firms. The ‘relationship with colleagues’ are used as a mediator to examine to what extent it explains the relationship between the independent variable “use of result control” and the dependent variable “work tension”. By using this model, we can achieve the goal to give managers from professional service firms more insight in the effect of applying result control on work tension.
Figure 1
Relationship with colleagues
Result control Work tension
H2: +
H1: +
2 Theory
In this chapter, I will explain some concepts that I use in this paper. I will start by describing what professional service firms are and in which they distinguish themselves from other firms. Then I will describe the use of management control systems and what kind of controls could be applied. Afterwards I will explain what results controls are, why it is needed and how to implement it. I have decided to use the social identity theory, because this theory is linked to our components of result control, the relationship with colleagues and work tension. This theory can provide me with results from previous research that I can continue to build on in my research. I expect that the social identity theory helps me to explain that through a better relationship with colleagues this indeed leads to a lower work tension when more results controls are applied in the organization. Therefore, I will develop the hypothesis and describe the relationship between result control and work tension. Then I will give more insight into the relation between result control and the relationship with colleagues. Finally, I will describe the relation between the relationship with colleagues and work tension.
2.1 Professional service firms
Based on the work of Auzair & Langfield-Smith (2005) and Von Nordenflycht (2010) the characteristics of professional service firms will be explained. According to Auzair & Langfield-Smith (2005), there are three types of service firms, namely: Professional services, mass services and service shops. Professional services are organizations with relatively few transactions, highly customized, process oriented, with relatively long contract time, with most value added in the front-office and where considerable judgment is applied in meeting customer needs (Auzair & Langfield-Smith, 2005, p. 402). You could think of management consultancies and corporate banking. In contrast to professional services, mass services have many customer transactions, involving limited contract time and little customization (e.g. tobacco, newspaper retailers and transport). Finally, service shops are a categorization that falls between professional and mass services (e.g. hotels or rental services). In this paper, I focus on the professional services firms.
Von Nordenflycht (2010, p. 160) distinguished three different characters with regard to professional service firms: (1) knowledge intensity, (2) low capital intensity and (3) professionalized workforce. Knowledge intensity is about the knowledge from the employees in the organization. This means that organizations are dependent of the employee’s knowledge. Low capital intensity means that the results that professional service firms deliver to customers
more depends on the input from employees and less on the use of (physical) capital. Professionalized workforce means that the organizations have employees in the organizations with the right attitude and the required knowledge. So, it is important for organizations to have the right people. This is because, in return for the service the professional service employees deliver, the organization which they work for, get paid. From an organizational perspective, it is important that professional service employees do their work good and within the set time. The success of professional service firms depends on their people and their resources (Fu et al., 2015, p. 329). In summary, we can conclude that employees are essential in professional service firms, given that on the basis of their knowledge and skills, the organization earns money.
2.2 Management control system
This study focuses on improving our understanding to optimal management control systems (hereinafter: MCS) design to improving performance outcomes (in this study: less work tension for employees, so that they can perform better). Therefore, it is important to describe the operation of the MCS to understand what it is and how it is used. Especially because the independent variable in this study: ‘result control,’ which will be explained in more detail later, is a part of the MCS.
Simons (1995) has the following definition for a MCS: “management control systems are the formal, information-based routines and procedures managers use to maintain or alter patterns in organizational activities”. The MCS exist of four formal, information-based systems which are different in their relationship to strategy and their use by top managers, these are belief systems; boundary systems, diagnostic control
systems and interactive control systems. These four levers of control are presented in figure 2. Belief systems are about the explicit set of shared beliefs that define basic values, purpose and mission (it’s about the core values of the organization). Boundary systems are formally stated limits and rules that must be respected, it’s about the risks to be avoided. Diagnostic control systems are feedback systems to monitor
the focus is on the critical performance indicators. Finally, the interactive control system is a control system that managers use to regularly and personally involve themselves in subordinates’ decisions. The focuses attention is on strategic uncertainties, leading to new initiatives and strategies.
In addition to the framework of Simons (1995), Merchant and van der Stede (2017) argue that MCS composed of three types of controls: (1) result controls, (2) action controls and (3) social controls. They describe that result control is about defining performance measures and reward good results. Results control is therefore good linked with the diagnostic system of the model of Simons (1995). Action control is about focusing on the employee’s behavior to ensure that employees perform actions beneficial to the organization. Action control and boundary systems (model of Simons 1995) are good in line with in each other, they are both focused on risks to be avoided. Social control is based on self-control and social control. Belief systems (Simons 1995) and social control are both about the core values of the organization and the employee. In summary, both the model of Merchant van der Stede (the three types of control) and the model of Simons (levers of control) give good indication about how you can control the input (by putting social controls or belief systems), the process (by introducing action controls or boundary systems) and the output (by using results controls, the diagnostic systems or interactive systems).
To be able to determine which type of control should be applied, figure 3 helps to determine which control should be used in which situation. The first question is if the organization is able to measure performance results. If not, when actions are observable, action controls should be used. When actions are not observable, social controls are desirable. When the organization is able to measure the performance results and able to know which actions are desirable, actions and / or results controls should be used. When measuring performance results is possible, but there is no knowledge about which actions are desirable, results controls should be used.
2.3 Result controls
Before starting with providing earlier results with regarding to result control and work tension and the role of the relationship with colleagues, it is important to define the variables properly. In prior subchapter, result controls have already been briefly mentioned and explained in comparison with other types of control. Now, it will be explained in more detail.
Merchant and van der Stede (2017) stated that result control is about measuring the results of the employees’ effort. To implement results controls, organizations have to define the dimension on which results are desired; measuring the performance in the chosen dimensions; setting performance targets for employees to attain for each of the measures; and providing rewards for target attainment to encourage the behaviors that will lead to the desired results (Merchant and van der Stede, 2017, p. 38). In fact, the organization holds employees accountable for the results they deliver. These delivered results will be compared with the targets that must be achieved by employees. As a result, employees become rewarded (punished) for favorable (poor) results when achieving (not achieving) those targets.
Result control is necessary because managers have to monitor their employees. Monitoring is needed, because managers delegate decision right towards their employees. Through the delegation of decision rights from managers and employees, a control problem arises.Result controls are therefore commonly used for controlling the behaviors of employees at many organizations (Merchant & van der Stede, 2017, p. 34). Further, Merchant and van der Stede (2017, p. 12) appoint three main categories for the causes of the needs of control: (1) lack of direction, (2) motivational problems and (3) personal limitations. Lack of direction have to do with that employees sometimes simply do not know what the organization wants from them. Motivational problems are common because individuals are self-interested and the organizational and individual objectives do not naturally coincide (2017, p. 12). When there are personal limitations then there are external circumstances (e.g. lack of training, stamina of the employee, etc.) excluding lack of direction or motivational that ensure that management control is needed. Before starting with implementing results controls, organizations have to realize that results controls work best only when all of the following conditions are present: (1) organizations can determine what results are desired in the areas being controlled; (2) the employee have significant influence on the results for which they are being held accountable; and (3) organizations can measure the results effectively (Merchant & van der Stede, 2017, p. 42). In sum, it is important to know that results control is only effective when the desired results can be influenced by the employees and the results are measurable.
2.4 Social Identity Theory
In this subsection, I will describe the social identity theory (SIT). Based on this theory, I will develop my hypothesis and therefore, some background information about this theory is needed to understand why the SIT contributes to the role of the relationship with colleagues.
The SIT was developed by two British social psychologists in 1979: Henri Tajfel and John Turner. An overview of this theory is given in figure 4. The purpose of the SIT is to specify and predict the circumstances under which individuals think of themselves as an individual or as a group member (Ellemers, 2017). With the SIT,
the psychologists propose that a person’s sense of who they are depends on the groups to which they belong (Tajfel and Turner, 1986). They suggest in their research that people might act different in varying social contexts, for example like their family, the sports team they follow or the neighborhood they live in. When people perceives themselves as part of a group, that is an ingroup for them. When they do not identify themselves with another comparable group, this is called an outgroup (David, 2015). Tajfel and Turner (1986)
describe three processes that can create an ingroup or outgroup mentality, these are: social categorization, social identification and social comparison. social categorization is about categorizing people in order to understand and identify them, it’s about which group you or another person belongs to (Stets & Burke, 2000). Social identification is about adopting the identity of the group that we belong to, and that we act in the way that we perceive members of that group act (Stets & Burke, 2000). After that a person categorize themselves within a group and identify themselves as being members of that group, we tend to compare our group, the ingroup, against another group, an outgroup (David, 2015).This is social comparison. In this paper, I will focus on the social identification and social comparison aspects from the SIT. Especially the social comparison component is important, because this is about the ‘us’ versus ‘them’ mentality. I expect that based on the SIT, when employees have a better relationship with their colleagues this leads to less work tension because when results controls are applied they feel that they work together as a team (ingroup) against the competition (outgroup). I think that the SIT is good to examine in professional service firms, because
competition and work pressure are very high in this industry (Kuang et al., 2014). So, this paper uses the SIT with regard to result control and the relationship with colleagues to examine the effect on work tension experienced by employees.
Prior research shows us that, based on the social identity theory, when people are in the same conditions (f.e. imagine employees are evaluated by result controls), they are more willing to help other person provide them with the four forms of support identified by House (1981) – instrumental, emotional, companionship and informational (Hopkins, 1997, p. 1232). In addition, Haslam (2004, p. 196) says that social identity provides a basis for effective support against stress, because less stress is experienced by the employee when employees want to help each other (Haslam, 2004, p. 196). Further, people who have a high extent of social identification reported lower levels of burnout than low social identifiers (Reicher & Haslam, 2003). Based on the above results, we could argue that the use of result control lead to a more ‘ingroup’ feeling and that less work tension is experienced because all employees want to perform better than the ‘outgroup’ and have therefore a higher level of social identification with each other. But social identification has also a downside. Preliminary evidence shows that social identification does have a role in burnout emerges (Haslam et al. 2004; Haslam et al., 2003). The research of Haslam et al. (2003) was done at a consultancy firm (part of professional service firms) and in this study, there was a significant negative relationship between social identification and a burnout (Haslam, 2004, p. 203). A burnout is seen as a consequence of long-term stress through a process of emotional exhaustion, depersonalization or dehumanization, and a belittling of personal accomplishment (Maslach & Jackson, 1981). In sum, social identification in relation to work tension has different outcomes in prior literature. But, how is this between result control and work tension?
2.5 Result control and work tension
In prior subchapter, the social identity theory and some preliminary results with regard to social identification and work tension were discussed. In this chapter, I will describe the relation between result control (which is already discussed in chapter 2.3) and work tension. Work tension is related to work related stress, which can be defined as a person’s individual response to environmental stressors (Vigoda, 2002, p. 574). These environmental stressors can have various causes, from private issues to work-related issues, stress can arise in people. In addition, Work tension or stress, is the degree to which the employees feel that they cannot perform work in the way they believe it should be done. (Dijkhoff, 2016). When a lot of work
tension is experienced, this is first of all bad for the employee, but on the short and long term also bad for the performance of the employee at work, which leads to a poorer performance of the company. The question is, does more use of result control lead to more work tension? Shields et al. (2000) investigated the effect of performance control systems on the firm performance with job-related stress as intervention variable. Shields et al. (2000) found that when standard-based incentives (f.e. bonuses when targets are met) are used, that his leads to a lower degree of job-related stress. The reason for this was found by Merchant and Manzoni (1989) that provides evidence for the fact that 80 to 90 percent of the targets will be achieved on time. Because goals are often achieved, this ensures a relaxed attitude for the employee when result control is applied. Thereby, the use of result control leads to clarity for the employee (and thus no lack of direction, see also 2.3), which removes stress. Besides that, Shields et al. (2000) found also that more standard tightness (f.e. goal difficulty, budgetary slack), which is defined as the amount of resources needed to perform at the level of a standard minus the amount of resources provided to perform (2000, p. 190), leads to more job-related stress. On top of that, the most important finding in the research of Shields et al. (2000) was that the positive effect of MCS on firm performance are diminished through the indirect effect of job-related stress. Also, Smith et al. (1992) found that employees (telecommunication workers) who were constantly electronically monitored at work perceived their working conditions as significantly more stressful and that performance of employees decreased. Tight targets therefore ensure a higher degree of work tension. In addition, also Crawford et al. (2010) found that high job demands are associated with increased burnouts. This insinuates that high job demands (f.e. tight results controls) leads to a higher work tension and more burnouts. In addition to above results, Xie & Schaubroeck (2001) argue that a job high in demands and low in control has a negative effect on human health. Whereas high job demands and a high level of control lead to employees who feel younger. This means that employees needed challenging goals to feel less work tension, but a high level of control is decisive. Thereby, Brouer & Harris (2007) found that individuals feel that they experience lower work tension when they have a high positive activity and high frequency of interaction with their supervisor. This implies that more interaction (and thus: control) between manager and employee reduces work tension for individuals. Thus, through the interaction with supervisors, the work tension of employees can decrease when result controls are used.
In sum, prior literature (Shields et al. 2000; Smith et al. 1992; Crawford et al., 2010) shows that through control systems work tension increased. Based on these finding, we could argue
that more result control leads to more work tension. In contrast, it is expected that with a high level of interaction (and control) with the supervisor (Brouer & Harris, 2007), work tension will be reduced. Based on Brouer & Harris (2007), we could argue that more result control lead to less work tension, but the risks of burnouts should not be underestimated (Crawford et al. 2010; Xie & Schaubroeck, 2001). Therefore, based on the above results, the following hypothesis is developed and will be examined in this paper:
Hypothesis 1: Result control has a positive relationship with work tension
2.6 Result control and relationship with colleagues
In the previous subsection, we discussed the direct relationship between result control and work tension. In this section, we will focus on the role of the relationship with colleagues. Eventually, it is expected in this paper, that when more results controls are used, employees through the social identification and social comparison (‘us’ versus ‘them’) have a better relationship with other colleagues and that therefore employees experience less work tension in their organization. The question is, do result control lead to a better or worse relationship with other employees?
Based on the social identity theory, when people feel social identify, they are motivated to provide mutually supportive coordinated contributions and also make up for any other group members limitations or shortcomings. This is for example, by helping them out, providing them with social support or doing extra work (Hopkins ,1997; Tyler, 1999). In addition, Levine et al. (2005) found that a person’s willingness to help a stranger in distress is enhanced when the stranger in question is perceived to share a social identity with the prospective helper. Further, Haslam (2004, p. 170) argue that sharing of labour can help to promote group productivity and with this he suggests that this also could lead to better relationship between employees. Based on these finding, we could argue that when results controls are more applied, employees feel more connected to each other. The reasoning for this argument is that when employees face with limitations (goals that are difficult to achieve), other colleagues will identify themselves with them and help them with doing extra work or giving social support. Therefore, we could argue that more use of result controls leads to more limitations (and shortcomings) to achieve the set targets for employees, but through social identification this leads to more social support to reach the targets and eventually to a better relationship with other colleagues.
The above reasoning is based on the ‘social identification’ part of the SIT. The next step after social identification is social comparison. Ashmore et al. (2001, p. 19) argue that an ingroup
positivity is enhanced by social comparison. Further, individuals value their ingroup positively and maintain positive, cooperative relationships with colleagues of their organization. (Ashmore et al., 2001). When results controls are applied, this can lead to an ingroup vs. outgroup mentality, where the employees of the ingroup are more attached to each other. This is because using result controls implies that the company want to perform better than before and that targets must be met. Targets are set, for employees individually and for the company as a whole, to perform better than the competitors. This competition drive between different firms lead to a more ‘ingroup-feeling’ rather than an ‘outgroup-feeling’. But, a possible downside of the use of results controls is the organization is that this can lead to internal competition. Being competitive has its advantages in terms of the better relationship with colleagues, as long as this is not at the expense of the work of the other colleagues. On the contrary, seen from utilitarianism, the majority of employees have a better relationship with colleagues because of the ‘us’ against ‘them’ mentality. So, not on only from the starting point of social identification, but also from social comparison, a better relationship with colleagues is expected by using the SIT when result control is applied.
Based on our reasoning above, result control can lead to a better relationship with colleagues. This is based on the work of Hopkins (1997) and Tyler (1999) who argue that when group members have limitations or shortcomings (this is often the case when results controls are used), other group members will be helping them out, by providing them with social support or doing extra work. In addition, Haslam (2004, p. 170) argue that sharing of labour can help to promote group productivity and with this he suggests that this also could lead to better relationship between employees. Finally, the results of Ashmore et al. (2001) give reason to suggest that through the ‘ingroup-feeling’ leads to a better relationship with colleagues. Based on these results, the following hypothesis will be investigated:
Hypothesis 2: Result control has a positive relationship with the relationship with colleagues.
2.7 Relationship with colleagues and work tension
In previous subsection, we discussed the relationship between result control and the relationship with colleagues. In this section, we will focus on the relationship between the relationship with colleagues and work tension.
In previous literature dealing with this topic, Haslam (2004, p. 196) describe that social identity provides a basis for effective support against stress, because less stress is experienced by the employee when employees want to help each other (Haslam, 2004, p. 196). He suggests that through social identification, employees are more willing to help each other, which leads to a better relationship with colleagues and less stress. This assumption is supported by Hopkins (1997) and Tyler (1999) who prove that when employees identify themselves with others, this could lead in a positive effect by helping other colleagues by providing them with social support or doing extra work. So, social identification has a positive side for effective support against stress and it seems that also the relationship with others (colleagues) will be better, because providing social support or doing extra work.
But social identification does not just lead to benefits. Preliminary evidence shows that social identification does have a role in burnout emerges (Haslam et al. 2004; Haslam et al., 2003). Carmona et al. (2006, p. 94) found that through social comparison employees who compare themselves with colleagues who are doing better, experienced relatively more burnout. But with regard to social identification, Reicher & Haslam (2003) argue that people with a high extent of social identification reported lower levels of burnouts than people with a low extent of social identification. The reason for this was driven by a sense of lack of accomplishment in comparison to others (Haslam, 2004, p. 203). Based on the above results, we could argue that through social identification, work tension decreased. But through social comparison, work tension increased. In addition, Brouer and Harris (2007) found that individuals feel that they have low work tension when they had high positive activity and high frequency of interaction with their supervisor. This implies that more interaction between manager and employee reduces work tension for individuals. The above reasoning says nothing about the relationship between employees, but it does show that the social aspect cannot be underestimated. In this study, it is expected that through social identification, employees have a better relationship with their colleagues and experience less work tension. Therefore, the following hypothesis is developed:
3 Method
In this chapter, I will describe the method that I use to get an answer to the research question. The structure of this chapter is as follows, first I will present the data and explain which respondents are included in the survey. Secondly, I will present the instruments and tell how they are used in this investigation. Finally, I will present a data-analysis of the instruments that were defined with the means, standard deviations of all independent, dependent, mediator and control variables.
3.1 Respondents
In this research, I use a survey which has been taken between November 2015 and January 2018. The purpose of this research is investigating the way in which professional service firms design their management accounting and control systems. Especially, they investigate the effect that these systems have on the professionals working in these organizations. In total, 730 participants, who work as professional service employees, have completed the survey. Professional services are organizations with relatively few transactions, highly customized, process oriented, with relatively long contract time, with most value added in the front-office and where considerable judgment is applied in meeting customer needs (Auzair & Langfield-Smith, 2005). The participants are working in a professional field, have more than three years of work experience (but not more than ten year) and work in organizations with more than fifty employees. The 730 participants, work in many different industries. Consultants (112) and employees working the accounting industry (105) are the groups that are most represented in this study.
In my research, I will use the answers of the professional service employees to questions related to the use of result control in their organization, the relationship with their colleagues and the degree of work tension they experience. The result of these questions will be analyzed to examine if the expected assumptions are true. The dependent variable in this research will be ‘work tension’, the independent variable will be ‘result control’, the mediator will be ‘relationship with colleagues’ and the control variables will be ‘gender’, ‘firm size’ and ‘firm ownership’.
3.2 Instrument
In this subsection, I will describe the instruments which will be used to answer the research question. First, I will use the dependent variable ‘work tension’ to measure whether the employees experience work tension. Second, I will do the same for the independent variable ‘result control’, to what extent is it used in the organization? Thirdly, I will do the same for the mediator variable ‘relationship with colleagues’. The survey that is used contains multiple questions which regard to these topics, the questions are involved in appendix 1, 2 and 3. Further, I will also involve some control variables in my analysis. I want to correct for the influence of gender, firm size and firm ownership.
3.3 Data-analysis
In this subsection, I will do a first data-analysis with the instruments that were defined in the previous subsection. Further, I will also present the means, standard deviations, and (Cronbach’s) alpha of all independent, dependent, mediator, and control variables. Finally, I will present all descriptive statistics and the correlation matrix.
3.3.1 Work tension
First, I use the dependent variable ‘work tension’ to measure whether the employees experience a lot or little work stress. The survey contains multiple questions that gave answer on the next question: to what extent do you experience work tension at work? In appendix 1, we can see which questions are involved. On a scale of 1 – 5, the respondents indicate to what extent they experience work tension. A higher score means that there is a ‘strong work tension’. A lower score means that there is a ‘weak work tension’.
When doing the reliability statistics in SPSS for all eight questions about work tension as presented in appendix 1, the output shows us a Cronbach’s Alpha of 0.762. This is good news for the research, because 0.762 is higher than the required rule of thumb of 0.7. If the question “The type and structure of my employment gives me the opportunity to fully express myself as a professional” is deleted, Cronbach’s Alpha will rise to 0.842. This is a fairly large increase, but not necessary since the rule of thumb of 0.7 has already been met. When doing the factor analysis (see appendix 6: Rotated Component Matrix), the first eight questions are all about ‘work tension’. The factor analysis there shows us that there are two factors and that indeed the question about: “The type and structure of my employment gives me the opportunity to fully express myself as a professional” falls outside the other questions (see appendix 6).
Therefore, we use the other questions to examine the extent of work tension experienced by the employees. The seven questions that are used can be found in column WT of Appendix 6. Also in table 1, the seven questions are given, with also the total variance explained (52,041%) and the Cronbach’s alpha (0,842) are included.
Table 1: Factor analysis for work tension
The 7 questions (see also column WT of A6) WT
1. I do not have enough time to complete my work the way that I think it should be done.
0,607
2. Due to a lack of adequate resources and materials, I cannot execute my assignments properly.
0,663
3. My organization hinders me from doing my work properly. 0,806 4. In my organization, there is a conflict between the work standards and
procedures of the organization and my own ability to act according to my professional judgement.
0,753
5. I have to alter my professional behavior in order to perform my job the way my organization wants me to.
0,630
6. In this organization, I can’t perform the way that I think I should. 0,730 7. I could do my job much better without the conditions imposed by my
organization.
0,733
Total Variance Explained (in %) 52,04
Cronbach’s alpha 0,842
3.3.2 Result control
Secondly, we use the independent variable ‘result control’ to measure whether the employees have much or little to do with result controls. The survey contains multiple questions that answer the next question: to what extent is result control used in the organization? Appendix 2 contains the questions that are asked to the respondents about the use of result control. On a scale of 1 – 5, the respondents indicate to what extent result control is used in their organization.
When doing the reliability statistics in SPSS for the eight questions about result control as presented in appendix 2, the output shows us a Cronbach’s Alpha of 0.712. This 0.712 is higher
than the required rule of thumb of 0.7. But it seems like the eight questions about result control can be divided in two groups. The first five questions in appendix 2 are about to what extent result control (RC) is used. The last three questions in appendix 2 are more about to what extend are subjective performance evaluations (SPE) important in the organization. When doing the factor analysis, the rotated component matrix tells that there are indeed two factors (see appendix 6: component matrix and also in table 2 here below). In table 2 we can see that Cronbach’s alpha of result control is 0,808 (which is good) and the total variance explained is 36,64%. Further we can see that the other factor, named: subjective performance evaluations (SPE), has a Cronbach’s alpha of 0,416 and a total variance explained of 16,63%. This is no problem for this study, because only the factor result control will be examined in the analysis.
Table 2: Factor analysis for result control
The 8 questions (see also column RC of A6) RC SPE
1. In my job, there is a performance measure for everything. 0,765 2. My organization sets a large number of performance goals/targets that I
am expected to meet.
0,722
3. Employee attainment of goals/targets is checked constantly. 0,801 4. My supervisor frequently checks to make sure that I am meeting my
performance targets.
0,873
5. In our organization, goals/targets are essentially a guideline rather than a true commitment.
0,668
6. My supervisor is very considerate of my explanations of deviations from pre-established goals/targets.
0,553
7. Responding to new, unforeseen opportunities is considered more important by my supervisor than achieving pre-established goals/targets.
0,765
8. In my organization, employees are expected to meet pre-established goals/targets with no expectations.
0,647
Total Variance Explained, when all 8 questions are included (in %) Total Variance Explained, when RC/SPE are generated separately, (in %)
36,64 56,76
16,63 46,40
3.3.3 Relationship with colleagues
Thirdly, I’m not only going to investigate the direct effect of the use of result control on ‘work tension’, but also the influence of the relationship with colleagues. In appendix 3, question 2 and 3 will be used to say something about the influence of the relationship with the colleagues. Question 2 is as follows: “I socialize with my colleagues outside of work.” Question 3 is: “I am not friends with any of my colleagues. Question 3: “I am not friends with any of my colleagues” is asked in a negative way. That is why this variable will be transformed so that here to, a higher score means that there is a good relationship with colleagues and a lower score means that there is a less good relationship with colleagues. Now, applies to both questions, a high score means that there is a ‘strong relationship’ while a low score means that there is a ‘weak relationship’ with other colleagues. The other questions in appendix 3 are excluded, because these questions are more about what the organization does for the employees instead of measuring the relationship with colleagues.
When doing the reliability statistics in SPSS for the two questions about the relationship with colleagues, the output shows us a Cronbach’s Alpha of 0.625. Unfortunately, this 0.625 is lower than the rule of thumb of 0.7. But a Cronbach’s Alpha that exceeds 0.5 is acceptable according to Maas (2009, p. 11). In addition, it is also very difficult to get a high Cronbach’s alpha with two components. Therefore, the two questions can be used in this survey to determine the relationship with colleagues.
When doing the factor analysis with all eight questions regarding the relationship with colleagues and the organization (see appendix 3), the component matrix tells that there are two factors (see appendix 6: Rotated Component Matrix). As expected, the two questions are indeed one factor in this analysis (see column: REL in appendix 6). The total variance explained here is 72,71%.
Table 3: Factor analysis for the relationship with colleagues
The 2 questions (see also column REL of A6) WT
1. I do not have enough time to complete my work the way that I think it should be done.
0,762
2. I am not friends with any of my colleagues -0,858
Total Variance Explained (in %) 72,71
3.3.4 Gender
An important factor that could affect the outcomes of the hypothesis is gender. Nelson & Burke (2002, p. 92) found that women are more sensitive to stress than men. Therefore, I want to correct for the effect of gender. In the survey, the question: “What is your gender?” is involved. Here is option 1: “male” and option 2: “female”. In the survey, a total of 628 respondents fill in this question. There are 410 men in the survey, which is 65,3% of all respondents. Besides, there are 218 women, which is 34,7% of the whole population in this survey.
3.3.5 Firm Size
The size of the organization can also have a significant influence on the final results of this research (Auzair & Langfield-Smith, 2005, p. 416). It is possible that more results controls are used in larger organizations and less in smaller organizations. Therefore, I want to use ‘Firm size’ as a control variable. I want to correct for the effect of the size of the organization. In the survey, there is one question about firm size. This question is: “How many people are employed by your entire company?” The possible answers to this question in the research are (1) less than 100, (2) more than 100 but less than 500, (3) more than 500 but less than 5000 and (4) more than 5000. In appendix 4, this question is included and displayed as it is in the survey.
3.3.6 Firm ownership
The last control variable that I want to use is “Firm Ownership”. The ownership structure of a firm can have a significant influence on the preparation of result controls and the difficulty to achieve these targets. King and Clarkson (2015) stated that the ownership structure is very important for choices that have to be made with regard to management control systems (and thus: result controls). Therefore, I want to control for the effect of firm ownership and want to make a distinguish between profit and non-profit organizations.
In the survey that I use, one question is asked about this subject. This question is included in appendix 5. This question is: “Which of the following best describes the ownership type of your organization? The possible answers are: My organization is primarily owned by (1) people who work in the organization (i.e. partnership), (2) people employed outside of the organization (i.e. shareholders, investors) or (3) no one. I work in a non-profit/public organization. Most organizations (322) are primarily owned by people who work in the organization (i.e. partnership). After that, most organizations (255) are primarily owned by people employed outside of the organization (i.e. shareholders, investors). The smallest group (85) are organizations who are primarily owned by no one, because these employees work in a
non-profit or public organization. In my analysis, I would like to distinguish between non-profit and non-profit organizations to control for the effect of ownership. Therefore, organizations who are primarily owned by (option 1) people who work in the organization and (option 2) people employed outside of the organization, be merged into one variable that stands for profit organizations (they get the value 1). The last group who is primarily owned by (option 3) no one, are the non-profit organizations (they get the value 0). This means that there are 577 employees who work for profit organizations in this survey and 85 employees work for non-profit organizations.
3.3.7 Descriptive statistics and correlations matrixes
Before creating the correlation tables, we first start with presenting the descriptive statistics of all variables together. This overview is shown in table 4. All variables are explained earlier. In summary, the work tension that is experienced by employees is lower than the average of 3,0. The use of result controls is around the average of 3,0 as experienced by the respondents. Respondents experience on average a good relationship with their colleagues (3,6262 > 3). Most respondents are men (65,29%). Most employees work in larger organizations and finally, 86,17% of the respondents work in profit organizations.
Table 4: Descriptive statistics
Variables N Minimum Maximum Mean Stan
Work tension (7) 639 1,00 5,00 2,4672 ,79868 Result Control (5) 630 1,00 5,00 2,8606 ,82106 Relationship (2) 626 1,00 5,00 3,6262 ,96584 Gender 628 0,00 1,00 ,6529 ,47644 Firm size 660 1,00 4,00 2,9091 1,06578 Firm Ownership 662 0,00 1,00 ,8617 ,33479 Valid N (listwise) 621
By examining the bivariate correlation, the correlation matrix with all variables has been done. The results of the correlations (Pearson and Spearman) are given in table 5: Correlation Matrix. According to the Pearson correlation, this preliminary analysis indicates that there is a significant positive correlation between result control and work tension (,133).Between result
control and the relationship with colleagues there is also a significant positive correlation (,110). Further, there is a significant negative correlation between the relationship with colleagues and work tension (-,160). These preliminary results seem to indicate that all hypotheses are accepted. But, this will be further analyzed in chapter 4: Results. In addition, the control variable gender has a significant negative correlation with the relationship with colleagues (-,122). Firm size has no significant relationship with the dependent (work tension), independent (result control) or mediation (relationship with colleagues), but it does have a significant negative correlation with firm ownership (-,078). Thereby, firm ownership, has a significant positive correlation with result control (,087). Finally, there is a significant positive correlation between gender and firm ownership (,102).
From the non-parametric Spearman correlation perspective, there are few differences to be observed with the pearson correlations. The significant correlations found between the variables are the same except for one. Now, from a Spearman correlation perspective, between firm size and firm ownership there is no significant correlation between these variables. Table 5: Correlation matrix
Work Tension
Result Control
Relationship
with col. Gender Firm Size
Firm Ownership Work Tension ,133** -,160** ,053 ,032 -0,067 Result Control 0,128** ,110** ,045 ,042 ,087* Relationship with col. -,146** ,111** -,122** ,047 -,041 Gender ,046 0,60 -,125** ,076 ,102* Firm Size ,047 ,060 ,039 ,070 -,078* Firm Ownership -,065 ,082* -,041 ,102* -,064
Pearson correlation appears above the diagonal, Spearman appear below the diagonal **. Correlation is significant at the 0.01 level (2-tailed).
4 Results
In this chapter, the results of the hypotheses will be presented. Per hypothesis will be described whether there is a significance relationship, if this is positive or negative, what the explanatory power is and if the hypothesis is accepted or rejected. Finally, at the end of this chapter, I will give a short summary of all results in the context of the research question.
First, hypothesis 1 predicts that result control has a positive relationship with work tension. In other words, when more results controls are used, employees experienced more work tension in their organization. Table 6 contains the regression with ‘work tension’ as dependent variable and ‘result control’ as independent variable, the control variables are also included. At first, we see that the p-value in table 6 is significant (,000). In addition, we can conclude that there is a positive effect since the beta of result control is positive (,139). Further, the explained variance on the dependent variable ‘work tension’ is 2,3% (Adjusted R Square). The control variables gender and firm size do have a positive effect on work tension, but they are not significant (,170 and ,719). Firm ownership does have a significant negative effect (,016) on work tension, this means that in profit-firms there is less work tension experienced and more work tension is experienced in non-profit firms. In summary, hypothesis 1 is accepted, there is a positive correlation between result control and work tension, but the explained variance (0,023) is unfortunately low. The fact that hypothesis 1 is accepted, means that when more results controls are applied in organization, employees experience a higher degree of work tension. Table 6: Linear regression analysis work tension
Model Beta (standardized) T-statistic P-value
(Constant) 13,391 ,000 Result Control ,135 3,386 ,001 Gender ,055 1,375 ,170 Firm Size ,014 ,360 ,719 Firm Ownership -,097 -2,415 ,016 R-square ,029 Adjusted R-square ,023 F 4,545 Significance F ,001
Hypothesis 2 predicted that result control has a positive relationship with the relationship with colleagues. This means that when more results controls are used, employees have a better relationship with their colleagues. Table 7 contains the output of the linear regression with ‘relationship with colleagues’ as dependent variable and ‘result control’ as independent variable. At first, we see that the p-value in table 7 is significant (,000). In addition, we can conclude that there is a positive effect since the beta of result control is positive (Beta = ,126). Further, gender has a significant negative effect on the relationship with colleagues (Beta = -,128 and P-value = ,001), this means that women (because the dummy variable is set as men are 1 and women are 0) experience a better relationship with their colleagues then men. Firm size and Firm ownership do have influence on the relationship with colleagues, but this influence is not significant. In total, hypothesis 2 is accepted, there is a positive correlation between result control and the relationship with colleagues, but the explained variance (,028) is very low. Hypothesis 2 is thus accepted, which means that when more result controls are used in the organization, the employees experience that they have a better relationship with their colleagues.
Table 7: Linear regression analysis relationship with colleagues
Model Beta (standardized) T-statistic P-value
(Constant) 16,803 ,000 Result Control ,126 3,165 ,002 Gender -,128 -3,198 ,001 Firm Size ,045 1,122 ,262 Firm Ownership -,035 -,876 ,381 R-square ,034 Adjusted R-square ,028 F 5,425 Significance F ,000
Hypothesis 3 predicted that the relationship with colleagues has a negative relationship with the work tension. This means that when employees have a better relationship with their colleagues, they experience less work tension. Or vice versa, when employees have a worse relationship with their colleagues, they experience more work tension. Table 8 contains the
output of the linear regression with ‘work tension’ as dependent variable and ‘result control’ and ‘relationship with colleagues’ as independent variables. The control variables were also included in the analysis. At first, we see that the p-value in table 8 is significant (,000). The negative beta of -,177 tells that there is a negative correlation of relationship with colleagues and work tension. This negative correlation is significant, because of the p-value is ,000. In addition, the effect of result control on work tension is also significant in this regression (p-value = ,000). Because the effect of result control on work tension is also significant, there is a partial mediation. The control variables gender and firm size don’t have significant influence on work tension. Firm ownership does have a significant negative effect (,009) on work tension, this means that in profit-firms there is less work tension experienced and more work tension is experienced in non-profit firms. Because the beta of the relationship with colleagues is negative (-0.177) with regard to work tension and their relationship is significant (p-value = ,000) hypothesis 3 is accepted. So, hypothesis 3 is accepted, but the explained variance (0,054) is very low. Accepting hypothesis 3 indicates that despite a good relationship with colleagues, work tension is still increasing when more result controls are applied, but these work tension experienced is lower than expected because of a good relationship with colleagues.
Table 8: Linear regression analysis work tension, including relationship with colleagues
Model Beta
(standardized)
T-statistic P-value
(Constant) 13,672 ,000
Result Control ,168 4,233 ,000
Relationship with colleagues -,177 -4,442 ,000
Gender ,032 ,792 ,429 Firm Size ,015 ,389 ,697 Firm Ownership -,104 -2,622 ,009 R-square ,061 Adjusted R-square ,054 F 8,024 Significance F ,000
In total, all hypotheses are accepted, which means that the assumed expectations have been fulfilled. On the basis of accepting hypothesis 1 it is concluded that more use of result control leads to more work tension among employees. Accepting hypothesis 2 implies that the conclusion that more result control leads to a better relationship with colleagues. Finally, hypothesis 3 is accepted and a partial mediation is concluded. This means that the relationship between result control and work tension is only partially explained by the relationship with colleagues. Accepting hypothesis 3 indicates that despite a good relationship with colleagues, work tension is still increasing when more result controls are applied, but these work tension experienced is lower than expected because of a good relationship with colleagues. This indicates that when more result controls are applied in the organization, employees still experience more work tension, but for employees with a worse relationship with colleagues experience it is more the case than for employees with a better relationship with colleagues. In the end, when more result controls are used, there is always more work tension experienced, but for employees with a better relationship with colleagues, this work tension is lower than expected.
5 Discussion
This study aims to examine if the relationship with colleague mediates the relationship between result control and work tension. Previous studies have shown that using (result) controls leads to more work tension (or job-related stress) among employees (Shields et al., 2000, p. 196; Ganster et al., 2011, p. 225). But in this research, we expect that when more result controls are used, employees through social identification and social comparison (‘us’ versus ‘them’) have a better relationship with other colleagues and that therefore employees experience less work tension in their organization. This research is a response to Shields et al. (2000) who ask for more research that improve our understanding of the effect of control systems on job related stress (or work tension).
In previous chapter, the results were discussed. The outcomes of these results are that (1) more use of results controls leads to more work tension, (2) more use of result control leads to a better relationship with colleagues and (3) a better relationship with colleagues leads to less work tension. All developed hypotheses are accepted. The expected assumption that a better relationship with colleagues reduces work tension is therefore true.
5.1 Theoretical implications
In this study, based on prior literature it is expected that by applying more result controls in professional service firms, employees experience a more ‘ingroup-feeling’ because of social comparison, which lead to employees who are more willing to help each other and therefore have a better relationship with other colleagues. This better relationship with colleagues will then be expected to lead to less work tension among employees. If this expectation is true, the relationship with other colleagues stated that less work tension is experienced when more result controls are used in professional service firms. To examine this assumption, the following research question in this study is developed: “How does the relationship with colleagues affect the relationship between result control and work tension in professional service firms?”
Based on the results in the study, we could say that the relationship with colleagues indeed affect the relationship between result control and work tension in professional service firms. Normally, applying more result controls leads to more work tension. But, when employees have a better relationship with colleagues, work tension will be reduced. A partial mediation has been concluded and shows us that a better relationship with colleagues indeed decreases
the work tension that is experienced by employees. Work tension is still increasing when more result controls are applied, but these work tension is lower than expected because of a good relationship with colleagues.
This study contributes to prior literature by responding to Shields et al. (2000), who ask for more research that improve our understanding of the effect of control systems on job related stress (= work tension). In addition, this study is an answer to the request from Auzair & Langfield-Smith (2005) who argue that future research has to be done in professional service firms. Thereby, managers now will better understand the effect of the use of result control on work tension and understand the role of the relationship from employees with their colleagues. It was expected that when managers adopt more results controls, employees have more work tension (Ganster et al. 2011), this study has shown that this is indeed the case, but that through a better relationship with colleagues, work tension will be reduced. This conclusion is an important contribution to current literature.
5.2 Limitations
The biggest limitation in this research is that the explanatory power in all hypotheses are very low. The explained variance in all hypothesis are respectively 2.3% for hypothesis 1, 2.8% for hypothesis 2 and 5,4% for hypothesis 3. This is a problem for external validity, because it makes it difficult to generalize these results to other groups. It may be possible that the results only apply to these respondents and not to a larger group of respondents. Further, a partial mediation has been concluded in this study, which is at the expense of the internal validity of this study. In addition, in this survey only two questions were used to determine the relationship with colleagues. This is very little, more questions about the relationship with colleagues would be better to have a good measuring instrument. Finally, this study concluded that a better relationship with colleagues leads to less work tension. Conversely, you could also say that less work tension leads to more socializing (and a better relationship) with colleagues. Less work tension leads then to more relaxed employees who have a better relationship with each other. This could be the case, on the other hand there is positive relationship between result control and relationship with colleagues.
5.3 Suggestions for future research
The low variance explained in all hypothesis make me aware of that there are other aspects that influence the work tension that is experienced by employees. It would be valuable to examine
the relationship with colleagues in relationship with work tension separately. Therefore, I would recommend investigating in other aspects that would reduce the work tension in a positive way. Managers will then have a better understanding of how they can reduce the work tension of employees in spite of using result controls in the organization. In addition, in this survey only two questions were used to determine the relationship with colleagues. This is of course very low, in the next study it is recommended to analyze several questions that have to do with the relationship with colleagues. Further, the positive relationship between result control and the relationship with colleagues is not obvious and therefore very interesting. More research on these variables can be a great added value. Finally, in this study the majority of the respondents are Dutch (71.2%). This makes it difficult to transform the results of this study to other countries in the world. It could be possible that Dutch employees are more stressful then employees in other countries. This also applies to the relationship with colleagues, are Dutch people good in having a good relationship with other colleagues in comparison with other countries? Now, we don’t know. Therefore, it is recommended to involve more employees from other countries in the investigation, allowing the transformation of results to other countries in the world.
5.4 Practical implications
In this subsection, I will discuss some practical implications. In response to the results of this research that through social identification and social comparison, work tension will be reduced, the organization can take actions to strengthen the relationship between employees. Firstly, what we do know now is that it is fine if people can get along well with each other on work. Through good relationships with colleagues, work tension can be reduced. Networking is fine, but obviously this should not be done too much. Favoritism within the organization is a real danger, so a relationship with colleagues that is to good has its disadvantages, this can endanger the integrity of the organization. Another risk is that a lot of social activity leads to a lower productivity of the employee. Now that we know that employees experience less work tension then expected due to a better relationship with their colleagues when result controls are applied, the organization (or managers) can take actions that can lead employees to have a better relationship with their colleagues. Possible actions that can be taken by the organization are planning team-building events for employees, hosts social event for employees or other social activities. This kind of actions can ensure that the work tension is eventually reduced.
