Gender and the impact of Control Tightness on the
Individual Performance of Professionals.
Name: Mohamad Yakub Mirkheel Student number: 10876006
Thesis supervisor: dr. ir. S.P. van Triest Date: 20 June 2016
Word count: 17.813
MSc Accountancy & Control: specialization Control Amsterdam Business School
2
Statement of originality
This document is written by Mohamad Yakub Mirkheel who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
3
Abstract
The sector of Professional Service Firms (PSFs) have been growing rapidly on an international level over the past decades and compose a great part of the current economy. This paper examines Management Control Systems (MCSs) Tightness in PSFs. The effect of Tightness in Cultural, Personnel, Behavior and Results Controls on the individual performance of professional male and female employees has been researched. The aim of this paper is to research if gender has a moderating effect on the relationship between Tightness and the individual performance of professional employees. The results show that Control Tightness in Cultural and Personnel Controls positively affect the individual performance of professionals. Tightness in Behavior Controls negatively affect the individual performance of professionals. The results also show that gender plays a significant moderating role on the effect of Control Tightness on the individual performance of professionals. The findings show that females perform better with tighter Personnel and Results Controls relative to males. Besides that males perform better with tighter Cultural Controls relative to females. This paper will also shed light on contingency variables and how these affect the MCS.
Preface
After finishing all the courses with a positive result, the last step which remained was writing my thesis. Before you lies my thesis, the crown on my higher academic education. I would like to thank Mrs. Helena Kloosterman MSc for giving me the opportunity to join her project, otherwise it would not be possible for me to do this research. I also want to thank my supervisor dr. ir. Sander P. van Triest for giving me valuable feedback and supervising me through writing my thesis.
4
Table of Contents
Contents
Statement of originality ... 2 Abstract ... 3 Preface ... 3 Table of Contents ... 4 Introduction ... 5Literature and Hypotheses development ... 8
Theory ... 8
Professional Service Firms ... 8
Management Control Systems ... 9
Gender ... 14
Research Methodology ... 18
Research Design ... 18
Variables ... 19
Statistical Analysis ... 20
Exploratory Factor Analysis ... 20
Preparation for hypotheses testing ... 24
Hypotheses testing ... 27
Results ... 30
Filtered analysis ... 33
Discussion & Conclusion ... 34
References ... 39
5
Introduction
In this paper we conduct research about the use of Management Control Systems (MCSs) in Professional Service Firms (PSFs). This paper will look at the effects of Control Tightness on the individual performance of professional male and female employees. This paper examines whether gender moderates the effect of Tightness on the individual performance of professional employees. By increasing our understanding of this relationship, firms may be able to make better choices in designing their MCS.
There has not been done a lot of research on the sector of PSFs (Von Nordenflycht 2010; Empson et al. 2015). Empson et al. (2015) mentions this lack of research on PSFs is due to the fact that PSFs are mostly private owned and are not legally obliged to present financial information, in contrast to governmental organizations and NGO’s. Von Nordenflycht (2010) also mentions that PSFs are of interest because they are presumed to be distinct from other type of firms i.e. PSFs face distinctive environments which demand distinctive theories of management. According to Von Nordenflycht (2010) PSFs are associated with three distinctive characteristics. Knowledge intensity, low capital intensity and professionalized workforce. Knowledge intensity indicates that the production of a firm’s output is based upon the input of complex knowledge. Low capital intensity indicates that the production of a firm’s output does not contain large amounts of nonhuman assets, such as equipment and inventories. Professionalized workforce indicates the features of professionalization, which are ideology and self-regulation. Because of these distinctive characteristics PSFs have makes doing research on PSFs interesting.
This paper will also give light to the subject of MCSs. According to Merchant and van der Stede (2012) MCSs are tools which help management and personnel to achieve organization goals. Cugueró-Escofet et al. (2012) mention that MCS are designed to achieve goal congruence, such that the goals of the employee are in accordance with the organizational goals, a tool for preventing agency conflicts. According to the theoretical framework of Merchant and van der Stede (2012) a MCS consists of the following types of controls: Results, Behavior, Personnel and Cultural Controls. Results Controls focus on the performance of professionals. They get rewarded upon good results and punished/not rewarded upon bad results. Behavior Controls focus on the behavior of professionals to ensure that professional behavior is in line with organization goals i.e. goal congruence. Personnel Controls focus on aspects such as improving professional skills. Cultural Controls focus on the corporate culture and the governance structure.
6 Besides the types of controls, the role of Control Tightness will also be addressed. Control Tightness is not explicitly defined in the literature (Chenhall 2003, 2010; Merchant and van der Stede 2012). In this research, Control Tightness has two forms; explicit tightness and implicit tightness. Explicit tightness is achieved by creating more controls, rules, regulations and procedures. The amount of tightness is achieved through selection of controls, the definition and the completeness. Implicit tightness is created by allowing less deviations from the targets set ex ante and the results. This tightness is created through decreasing the level of tolerance for deviating from the MCS.
Many research has been done on the behavioral differences regarding gender (Chodorow 1978; Twale & Shannon 1996; Cron et al. 2006; Bekker & Assen 2008; Bhagat & Williams 2008; Borghans et al. 2009; Bowie 2010; Thomas 2011; Dawson & Henley 2015), from these research we know that there is a psychological difference between males and females. Based on previous research (Twale & Shannon 1996; Cron et al. 2006; Bhagat & Williams 2008) we know for a fact that differences in male and female behavior also take place on an organizational level. Studies such as Twale & Shannon (1996) and Cron et al. (2006) and Bhagat & Williams (2008) have shown that there is a difference between male and female behavior in (the organizational culture of) professional service firms. The aim of this paper is to research gender differences regarding Control Tightness and whether this affects the individual performance of the professional in a positive or negative way. Besides that, this paper will look at the traits of male and female behavior and how this relates to the four types of controls in the MCS.
This research will also contribute to the existing literature there is on the subject of MCSs in PSFs. There have not been many studies on MCSs in PSFs, this is due to the fact that PSFs have a special structure (Von Nordenflycht 2010; Empson et al 2015). This paper examines the effect of MCS design on the individual performance of professional employees. Previous research has been done on performance improvement in PSFs (Stumpf, Doh & Clark 2002; Henri & Journeault 2010; Sakka, Barki & Coté 2013; Pernot & Roodhooft 2014; King & Clarkson 2015). The main focus of these studies were on the improvement of organizational performance, and not on the improvement of individual performance of professionals. Another contribution is the fact that this study will also focus on a psychological factor. The relationship between male and female behavior and the individual performance of a professional employee has not yet been researched before. This paper will also pay attention to the different behavior between male and female professionals in PSFs, which is a call for further research from Twale & Shannon (1996), Cron et al. (2006) and Bhagat & Williams (2008)
7 To gather this information a survey is being used. The survey consists of questions which address the following variables; Control variables, MCS variables, PSF characteristics and Outcome variables. Control variables (contingency variables) such as technology, environment, and size and structure. MCS variables such as Behavior, Results, Cultural and Personnel Controls (the four types of controls), for each type of control within the survey explicit and implicit tightness can be measured. In this research we will not make a distinction between implicit tightness and explicit tightness, as this is not our dependent variable. PSF characteristics such as task complexity, professionalized workforce, customer involvement, knowledge intensity and capital intensity. There are two outcome variables in the survey, professional tension and performance. Performance is addressed in two ways. The first one is unit performance, which addresses the performance of the organization as a whole. The second one is individual performance, which addresses the performance of the professional only. For this research the latter will be used, because of the fact that this research is about the individual performance of professionals. Professionals with different job occupations, age, gender, nationality et cetera have completed the survey.
Summed up, this paper will look at the effect of tightness in the four types of Controls (Behavior, Results, Cultural and Personnel) on the individual performance of male and female professionals in PSFs.
We have tried to examine this by asking the following question: “Does gender influence the relationship between Control Tightness and the Individual Performance of professionals?” The findings of this study have shown that Tightness in Cultural and Personnel Controls does positively affect the individual performance of professionals. However, Tightness in Behavior Controls is negatively related with the individual performance of professionals. Besides that the study found that gender plays a significant moderating role on the effect of Control Tightness on the individual performance of professionals.
The remainder of this paper is as follows. In the following chapter, literature on MCS and PSFs is reviewed and the hypotheses for this paper are developed. The third chapter will focus on the research design and the methods used for statistical analysis. Results and findings will be discussed in the fourth chapter. In the last chapter a discussion and the conclusions will be drawn.
8
Literature and Hypotheses development
Theory
Professional Service Firms
A PSF is a firm which delivers services using professionals. The PSF sector has grown over the past decades as one of the most profitable and significant sectors in the world (Empson et al. 2015). A few examples of big internationally renowned PSFs are the ‘Big Four’ companies e.g. PricewaterhouseCoopers, Deloitte, KPMG and Ernst & Young. Hill & Neeley (1988) define services as a broad and diverse range of products with the following characteristics; intangible, inseparable of production and consumption, difficulty of standardization and perishability. Services are intangible because they cannot be felt, relative to goods which can be touched and are tangible. Services are inseparable of production and consumption because they cannot be stored like goods. Services are difficult to standardize because they may vary relative to each of the customers’ needs. Services are difficult of perishability, like an empty seat on a flight. There are many forms of PSFs (Hill & Neeley 1988). PSFs such as audit firms, consultancy firms and law firms (Hill & Neeley 1988; Von Nordenflycht 2010; Empson et al. 2015). PSFs distinguish themselves from other types of firms through their ability to attract, mobilize, develop and transform the expertise and knowledge of the professional to create value for their customers (Lowendahl et al. 2001).
PSFs that use professionals, consist of employees with the right knowledge acquired through formal higher education (Hill & Neeley 1988). A professional can also be defined, according to Empson et al. (2015), as a person who has mastered a specific expertise or knowledge base. Professionals have many norms and values, such as the responsibility to safeguard the interests of customers and society as a whole (Empson et al. 2015). Besides that, professionals adhere to autonomy. Autonomy can be formulated as self-governance (Chodorow 1978).
Although it is clear what PSFs do, it can be difficult to purchase a service, compared to a product. A reason for this, is because service level agreements between PSFs and their clients are usually not clear due to the fact that customers do not have a precise specification of the agreement (Sonmez & Moorhouse 2010), compared to a car for example, the customers knows the exact specifications of the product. This uncertainty also brings a certain risk along with it, customers do not exactly know what they can expect. Customers have more risk with purchasing services compared to goods, whether these are individual clients or businesses.
9
Management Control Systems
Many studies have been done on the subject of MCSs (Simons 1987; Abernethy et al. 1997; Chenhall 2003; Henri 2006; Morris et al. 2006; Mundy 2010). According to Simons (1987) MCSs are formalized procedures and systems implemented to use information to change or keep patterns in an organizational activity consistent.
Management Control can be defined as “the process by which managers assure that resources are obtained and used effectively and efficiently in the accomplishment of the organization’s
objectives” (Anthony 1965). Merchant and van der Stede (2012) address management control as the process to influence the decision making behavior of managers. MCSs can be defined as the system which gathers management information which can then be used by managers to make decision which should lead to obtaining organization’s goals (Merchant & van der Stede 2012). Merchant and van der Stede (2012) mention four types of Controls in the MCSs. The four types of Controls are Behavior, Results, Personnel and Cultural Controls. Behavior Controls, also called action controls, focus on goal congruence. Behavior Controls can be addressed in different forms, such as behavioral constraints, pre-action reviews, action accountability or redundancy. Each of them addresses a different issue. Behavioral constraints address motivational issues. Pre-action reviews and Pre-action accountability address motivational issues, lack of direction and
personnel issues. Redundancy addresses motivational and Personnel issues. Results Controls aim at influencing the employee’s performance. Employees (or business units) with the highest performance receive the highest reward. This is only possible if the results can be measured in an effective way. Results Controls are very effective in motivating employees in achieving better results. Better results lead to better rewards. Personnel Controls aim at career ambitions and skills. Merchant and van der Stede (2012) mention that Personnel Controls focus on the natural tendency of employees to control themselves. Cultural Controls aim at the corporate culture of an organization. Cultural Controls focus on mutual monitoring and group forming with the same goals as the organization. This way pressure will be felt on an employee who deviates from the organizations goals.
Studies such as Gani et al. (2012) have shown that MCSs are related to performance. Gani et al. (2012) have researched whether strategy alignment with the MCS can lead to better financial and non-financial performance. Gani et al. (2012) address the misfit between strategy and the MCSs as the degree to which MCSs deviate from the optimal configuration for a given type of business strategy. Gani et al. (2012) address these business strategies according to the literature of Miles and Snow’s (1978). Miles and Snow’s (1978) present three strategies, prospector, defenders and
10 analyzers strategy. Prospector strategy, focuses on innovating and trying to change the market. Defenders strategy, focuses on the current market and is not focused on innovating. Analyzers strategy is a mix of the prospectors and defenders strategy. Through a survey, Gani et al. (2012) measured the types of strategy and the types of MCSs which was implemented by firms. The survey also measured performance. This way Gani et al. (2012) compared the performance of firms which did have a misfit and firms which did not have a misfit. Using a sample of 109 firms, Gani et al. (2012) have found that a misfit between strategy and the MCS is negatively related with both financial and non-financial performance. This study shows that the way MCSs are being designed does impact performance. Which is also discussed in literature regarding the contingency theory.
Studies in the past (Tiessen & Waterhouse, 1987; Otley, 1980; Chenhall, 2003; Stewart, 2010; Yan, 2011; Pernot & Roodhooft 2014) have shown that variables such as environment, culture, technology, structure and size have influence on the design of a MCS. These variables, which influence the MCS design are addressed by the contingency theory. Yan (2011) implies that “the contingency theory argues that organization structure and control systems design depends on contextual factors existing within the organization’s environment”. According to Pernot & Roodhooft (2014) it is clear that MCS are contingent but it is not clear whether contingency variables have any influence on the (organizational and individual) performance of an
organization. Pernot & Roodhooft (2014) have tried to research this by researching whether a proper MCS design of supplier relationships (the relationship between a manufacturer and a supplier) is associated with good or poor performance. This is done by examining the effects of MCSs misfit on performance. This misfit is measured by the degree of misfit between
contingencies, risk and performance. Pernot & Roodhooft (2014) have found that a MCS
contingency misfit is related with poor performance, however this misfit is only temporal because a manufacturer can adapt the MCS to fit the supplier relationship. One of the contingency
variables on which the contingency theory is based is culture (Chenhall 2011). According to Chenhall (2011), culture has been becoming an important factor in the design of MCS over the past 20 years. Henri (2006) and Chenhall (2003; 2011) mention that culture is an important factor which influence the design of a MCS. Chenhall (2003) mentions that all MCS contingency-based research have used the values of Hofstede (1984) to study the influence of culture. The approach of Hofstede (1984) is a national approach to culture and not an organizational approach. Thus the national approach of Hofstede (1984) will not be taken into consideration. In this research we will look at contingency (Control) variables such as level of education, size and experience.
11 Tightness in MCSs are defined in different ways. Like mentioned before Control Tightness is not explicitly defined in the literature (Chenhall 2003, 2010; Merchant and van der Stede 2012; Van der Stede 2001). According to Morris et al. (2006) MCSs have two functions, MCSs are implemented to induce desired Behavior or limit dysfunctional behavior. This means that the controls are tighter, there is less to no deviation from the desired behavior. This point of view fits the agency theory, which will be explained hereafter. Van der Stede (2001) mentions that tightness in controls can be viewed as little to no deviation from the objectives set ex ante. Merchant (1985b, 1998) addresses tightness in control systems as a system which assures that employees act as the organization wishes i.e. there are no agency conflicts. Tightening these controls can be done as follows. Defining goals such that they are according to the organizations goals. Communicating in such a way that employees accept organizational objectives. Monitoring more frequently, more detailed and timelier. And, rewarding more with better performance. Loose Controls however, are defined as the opposite of tight controls.
Tightness in controls brings a certain limitation to the amount of freedom an employee has (Van der Stede 2001). Tightness and looseness in controls are addressed in different forms/definitions in literature e.g. formal controls, bureaucratic control and organic controls. Formal/bureaucratic controls are controls with more rules and standardized procedures i.e. the controls are tighter. Organic controls are less bureaucratic, with fewer rules and standardized procedures i.e. the controls are looser (Chenhall 2003). The tighter the controls, the tighter the rules and regulations, which affects the amount of freedom an employee has (Van der Stede 2001).
The agency theory discusses the conflict between an agent (professional) and the principal (employer) (Eisenhardt 1989). Two problems arise from this agency conflict; moral hazard and information asymmetry. Moral hazard occurs when an agent takes more risk because he does not bare the risk of the principal. Information asymmetry arises when an agent has more information than the principal (Shapiro and Susan, 2005; Eisenhardt 1989). By using a MCS an organization can accomplish goal congruence among employees (Morris et al. 2006; Cugueró-Escofet et al. 2012), which means there is less deviation from the preferred behavior, which is a sign of tight controls. Which is also how Merchant (1985b, 1998) and Merchant & Van der Stede (2012) addresses tightness in controls. So, from a firm point of view Control Tightness is positively related with the individual performance of an employee. Another research which verifies the positive relation between Control Tightness and performance is the research of Chow (1983), although these were not professionals but car manufacturing employees. Other research on the subject of MCS has shown that employees are more committed, and that the MCS may be the
12 most useful, when formal (tight) and organic (loose) Controls are used together (Chenhall et al. 2010).
On the other hand, Mundy (2010) mentions the roles of MCS as complementary and interdependent. The goal of implementing a MCS is to exert control over achieving
organizational goals and to enable employees to search for opportunities and to solve problems (Mundy 2010). So, the focus of the MCS lies more on enabling the employees in doing their job. This means that the controls are less tight, the MCS enables employees to create opportunities. Besides that the findings of Raelin (1985) have shown that professional employees adhere to autonomy and independence when it comes to their individual performance. Believing that tight controls are negatively related with the performance of employees Raelin (1985) researched why professionals resist managerial control. The study of Raelin (1985) examines five aspects of a Professionals life which explains why professionals resist managerial Control. The five aspects are autonomy, overspecialization of technical skills, overemphasis on professional standards of evaluation, lack of interest in real-world practice, and disregard of organizational procedures (Raelin 1985).
Professionals wish to make their own decisions without pressure from external parties such as managers and outsiders (Raelin 1985), this is why professionals adhere to autonomy. Through the studying of higher education and (relatively low) socializing experiences, professionals become overspecialized in their technical skills, believing they do not need managerial Control (Raelin 1985). Professionals also believe that only their peers are adequate evaluators for their work, but they prefer their workplace free from regulations and/or interference of others, besides that professionals desire to be evaluated upon their results instead of their conformity to rules and regulations of the firm (Raelin 1985). Professionals have a lack of interest in real-world practice due to the fact that they have difficulties to adjust to the organizational life, because they see the organizational life as an intrusion into their practice, which is especially the case for young professionals (fresh from school) (Raelin 1985). Because professionals are individualistic by nature, and resist adjusting to rules and regulations they disregard organizational procedures (Raelin 1985).
Weighing both opinions. From a firm point of view formality and tight controls lead to better performance. From the viewpoint of a professional more autonomy and loose controls lead to better performance. The latter weighs heavier, because this is how professionals experience tightness (Raelin 1985) and our research is based upon a survey conducted by professionals. This
13 leads to the development of the first hypothesis. We assume that there is a negative relationship between Control Tightness and the Individual Performance of professionals.
The first hypothesis will show if there is a positive or negative relationship between Control Tightness in the four types of controls and the individual performance of professionals. This hypothesis will be drawn based upon the assumption that professionals adhere to autonomy, less formality, rules and controls (Raelin 1985).
H1a: The relationship between Tightness in Cultural Controls and the individual performance of professionals is negative.
H1b: The relationship between Tightness in Personnel Controls and the individual performance of professionals is negative.
H1c: The relationship between Tightness in Behavior Controls and the individual performance of professionals is negative.
H1d: The relationship between Tightness in Results Controls and the individual performance of professionals is negative.
The following graph gives a representation of the first hypothesis.
Employees Employees Employees Employees P E R F O R M A N C E
Loose controls Tight Results controls
P E R F O R M A N C E
Loose controls Tight Personnel controls
P E R F O R M A N C E
Loose controls Tight Behavior controls P E R F O R M A N C E
15
Gender
Many research has been done on the differences in gender Behavior. Like mentioned before, from previous research we know that male and female Behavior is different from a psychological point of view (Chodorow 1978; Twale & Shannon 1996; Cron et al. 2006; Bekker & Assen 2008; Bhagat & Williams 2008; Borghans et al. 2009; Bowie 2010; Thomas 2011; Dawson & Henley 2015). The following studies mention the difference between male and female behavior on a psychological level.
Bowie (2010) examined the relationship between childhood emotion regulation (to what extent can one regulate/handle their emotions), aggression and adolescent deviant social behavior. Bowie (2010) found gender differences in aggression levels and emotion regulation; for females lower emotion regulation lead to aggression at an adolescent age relative to males. Bhagat & Williams (2008) researched gender differences in relationships. They examined how both genders perceived the relationship with their psychologist. Bhagat & Williams (2008) found that females are more likely to show a psychologically or emotionally investment in a relationship relative to males. Besides that Bhagat & Williams (2008) also found that for females’ structural bonding (commitment coming from extrinsic motivators) has no significant effect on the relationship with the physician relative to males.
Borghans et al. (2009) researched gender differences in risk aversion. According to Borghans et al. (2009) their research provides “fresh insight into the relationship between psychological traits and economic preference parameters”. The findings of Borghans et al. (2009) show that females are more risk averse compared to males. The findings of Dawson & Henley (2015) have also shown that females view risk less positively compared to males i.e. females are more risk averse compared to males.
There are also differences between male and females regarding competitiveness. Studies show that males are more competitive relative to females (Hibbard & Buhrmester 2010; Buser et al. 2014). Buser et al. (2014) examined through career choices the level of competitiveness for each gender. Using a sample of 397 students, Buser et al. (2014) measured competitiveness through career choices. The findings show that males choose significantly more prestigious academic tracks (more math and science intensive) relative to females. Hibbard & Buhrmester (2010) also researched gender differences regarding competitiveness. Competitiveness was measured through two types. Competing to win i.e. competing to dominate others and competing to excel i.e. competing to surpass personal goals. A sample of 110 students show that males score higher regarding competitiveness to dominate others, there was no significance difference in competing to surpass personal goals.
16 Research has also examined the relationship between gender and cooperation. Sell & Wilson (1991), Brown-Kruse & Hummels (1993), Solow & Kirkwood (2002) and Irwin, Edwards & Tamburello (2015) have shown that males are more positively associated with cooperation relative to females. Research from Sibley, Senn and Epanchin (1968), Tedeschi, Hiester & Gahagan (1969) and Molina et al. (2013) contradicts this, they have found that males are
negatively associated with levels of cooperation relative to females. On the other hand, there are also studies (Dawes et al. 1977; Orbell et al. 1994) which have not found any significant
difference in levels of cooperation between males and females. For this research we will use the findings that support the assumption that females are more positively related to cooperation, relative to males. This is due to the fact that males are more autonomous relative to females (Chodorow 1978; Twale & Shannon 1996; Bekker & Assen 2008) i.e. males prefer working alone, which may lead to issues when cooperating. Another reason is that males are more competitive relative to females (Hibbard & Buhrmester 2010; Westbrook et al. 2011; Buser et al. 2014), which may also lead to issues when cooperating.
These differences in behavior also take place on an organizational level (Twale &Shannon 1996; Cron et al. 2006; Bhagat & Williams 2008). The following studies show the difference between genders in a professional working environment.
Twale & Shannon (1996) have found that males have more experience in the field relative to females, which is also found by Cron et al. (2006). Cron et al. (2006) researched the effect of gender in a professional service firm. Based upon three moderating factors: professional experience, motivation for becoming an entrepreneur and amount of effort put in the enterprise. Findings show that females have less experience as business providers relative to males. Females are less motivated by money relative to males e.g. females are willing to trade-off income for flexibility. Males are more motivated by income, as a reason for becoming a professional service provider.
The study of Chodorow (1978) found lower levels of autonomy for females relative to males. Bekker & Assen (2008) also measured gender differences in autonomy -connectedness. Autonomy-connectedness is defined by Bekker & Assen (2008) as “the need and capacity for self-reliance and independence”. Bekker & Assen (2008) measure autonomy-connectedness upon three subscales. The first one is self-awareness, the second one is sensitivity to others the last one is the capacity to manage new situations. The three subscales measure autonomy-connectedness. The findings show that females are more sensitive to other people relative to males. Another finding is that males are more positively related with self-awareness and the capacity to manage
17 new situations relative to females. Overall females scored lower on the autonomy subscales relative to males.
Other research which has been done on the behavior of professionals in a service firm is the study of Twale & Shannon (1996). Twale & Shannon (1996) researched gender differences in a professional service educational administration faculty. Twale & Shannon (1996) found that males have a more positive relationship with influencing departmental tenure, promotion decisions and valuing their autonomy. Overall the findings of Chodorow (1978), Twale & Shannon (1996) and Bekker & Assen (2008) show that males are more positively related with autonomy relative to females. This leads us to the formulation of the second hypothesis.
Like mentioned before, males are more competitive relative to females (Hibbard & Buhrmester 2010; Buser et al. 2014). The study of Westbrook et al. (2011) also examined competitive differences, however this sample was taken from adults in a working environment. A sample of 163 working adults show that males are more competitive relative to females.
The second hypothesis will address the relation between the tightness of the four types of control and the individual performance of male and female professionals. Like mentioned before,
according to Merchant and van der Stede (2012) there are four types of controls. The four types of controls are Behavior, Results, Personnel and Cultural Controls. Behavior Controls focus on goal congruence. Results Controls aim at influencing the employee’s performance. Personnel Controls aim at career ambitions and skills. Cultural Controls aim at corporate culture. Assuming that there will be different outcomes and combinations of individual performance in Behavior, Results, Personnel and Cultural Controls for each gender, we expect that males and females will react more positive/negative to the Controls they “relate to” compared to the Controls they do not “relate to”.
The main differences between the two genders, which mainly affect on how each gender relates to Control Tightness, are risk aversion and autonomy. Males are more likely to be risky compared to females (Borghans et al. 2009; Dawson & Henley 2015). This gives reason to assume that males are more likely to perform better with looser controls. Besides that, males are more likely to be more autonomous compared to females (Chodorow 1978; Twale & Shannon 1996; Bekker & Assen 2008), this also gives reason to assume that males are more likely to perform better with tighter controls.
Results Controls are measured in the survey by questions addressing measurements of
performance of the organization. Cron et al. (2006) have shown that males are more motivated by money relative to females, besides that females are willing to sacrifice income for flexibility.
18 Twale & Shannon (1996) have shown that males are more attracted to being promoted, relative to females. We assume that these characteristics will help males perform better with
measurements of performance, relative to females. Thus we hypothesize that males score a higher individual performance with looser Results Controls, relative to females.
Personnel Controls are measured in the survey by questions addressing the hiring process of the organization. Twale & Shannon (1996) and Cron et al. (2006) have found that males have more experience in the field relative to females. Besides that males are competitive relative to females (Hibbard & Buhrmester 2010; Westbrook et al. 2011; Buser et al. 2014). We assume that these characteristics will help males perform better in the hiring process, relative to females. So we hypothesize that males also score higher on individual performance with looser Personnel Controls, relative to females. (Twale & Shannon 1996).
Behavior Controls are measured in the survey by questions addressing rules and procedures of the organization. Studies such as Chodorow (1978), Twale & Shannon (1996) and Bekker & Assen (2008) have found that females are less autonomous relative to males. Besides that Cron et al. (2006) have found that males are more motivated to become an entrepreneur. We assume that these characteristics will help females perform better in behaving according to the rules and procedures of an organization, relative to males. So we hypothesize that females score higher on individual performance with tighter Behavior Controls, relative to males.
Cultural Controls are measured in the survey by questions addressing the relationship of the professional with the organization and colleagues. We hypothesize that females score higher on individual performance with tighter Cultural Controls, relative to males, because they are more cooperative (Sibley, Senn and Epanchin 1968; Tedeschi, Hiester & Gahagan 1969), less autonomous (Chodorow 1978; Twale & Shannon 1996; Bekker & Assen 2008) and less competitive (Hibbard & Buhrmester 2010; Westbrook et al. 2011; Buser et al. 2014) and thus more group oriented, relative to males. We assume that these characteristics will help females in performing better in having a relationship with the organization and colleagues, relative to males.
H2a: The relationship between Tightness in Cultural Controls and the individual performance of professionals is less positive for males relative to females.
H2b: The relationship between Tightness in Personnel Controls and the individual performance of professionals is less positive for females relative to males.
19 H2c: The relationship between Tightness in Behavior Controls and the individual performance of professionals is less positive for males relative to females.
H2d: The relationship between Tightness in Results Controls and the individual performance of professionals is less positive for females relative to males.
The following graph gives a representation of the second hypothesis.
Research Methodology
Research Design
In this paper we conduct a quantitative research. This is done by doing surveys in PSFs, such as banks, lawyer’s offices, healthcare organizations and audit firms. The respondents will differ from age, gender, country and the type of PSF they work in. The surveys will be spread through
contacts who work in PSF, and who meet the requirements.
The respondents need to meet the following requirements to conduct the survey. The respondent has worked in the field for more than 3 years (but probably less than 10 years). The respondent is not an owner/partner or board member of the organization. In other words, the respondent needs to be subject to the management accounting and control system rather than designing it. The respondent works in a medium/large sized organization i.e. the organization has more than
Female Female
Overall effect Overall effect
Male Male
Female Female
Overall effect Overall effect
Male Male P e r f o r m a n c e Loose controls P e r f o r m a n c e Loose controls Tight Cultural Controls
Tight Personnel controls
Tight Results controls
Tight Behavior controls P e r f o r m a n c e Loose controls P e r f o r m a n c e Loose controls
20 50 employees. The last requirement is that the respondent must speak and understand the
English language at a business level.
Variables
The survey contains control variables (contingency variables) such as technology, size, environment and country. The survey also contains PSF characteristic, MCS variables and outcome variables e.g. professional tension and performance (unit and individual performance). All the variables which are being used in the survey are shown in the appendix.
In this research, Control Tightness will be used as the independent variable. Like mentioned before, the survey measures tightness through two ways, explicit and implicit tightness. The results of the survey will give information about the explicit and implicit tightness of the four types of control (Behavior, Results, Personnel and Cultural). In this research we will not make a distinction between both types of tightness. The survey has two outcome variables which are professional tension and performance. Performance is measured in two ways, unit and individual performance. For this paper the latter will be used, individual performance will be the dependent variable. Individual performance is measured in two ways in the survey. “Individual performance in-role” is based on how respondents answered questions about their own performance assuming they were their own supervisor. The second one is “general individual performance”, the respondents answered these question about their own performance in general. Gender will be used as a moderating variable.
The following framework gives an overview of the variables and their relationship.
21
Statistical Analysis
The amount of surveys which have been conducted are 374. This is the ‘raw data’, because of the fact that this data cannot yet be used for statistical analysis. To work with proper data there was a need of filtering the data. Respondents who did not fill in the survey and/or left too many questions open were deleted, a percentage of 10% missing fields was maintained. This was a total of 58 surveys. Besides that the data has been checked on standard deviation, this way we can see if respondents filled in the same answer for the whole survey, a standard deviation of 0,5 was maintained. There were no respondents below the maintained standard deviation. The total number of remaining respondents is 316. This data consists of 193 males (61%) and 123 females (39%), with an average age of 35. From these respondents 68% have a Dutch nationality, which is the majority of the respondents.
The following respondents did not meet the rest of the requirements. Respondents with less than 3 years of working experience in the field. Respondents who are not a subject to the MCS; Chief Financial Officers, Head of Departments and Directors. These respondents were not deleted from the data, new variables were made up to filter these respondents. We will do the statistical analysis without the filtering and afterwards with the filtering. This way we can look if they make a difference in the results we get.
H1a Tightness in Cultural controls H1b Tightness in Personnel controls H1c Tightness in Behavior controls H1d Tightness in Results controls F= Female M=Male
I
n
d
i
v
i
d
u
a
l
P
e
r
f
o
r
m
a
n
c
e
F+ & M- F- & M+ F+ & M- F- & M+ H2a H2b H2c H2d
22
Exploratory Factor Analysis
According to Field (2014) exploratory factor analysis (EFA) can be used to prepare raw data for further analysis. EFA is an interdependence technique. That reduces the number of variables and gives structure to the data. The requirements of doing an EFA is to have a quantitative variables and the amount of observations is much larger relative to the variables, which is the case in this research. By doing the EFA the total number of factors are 14, this can be found in the appendix. The 14 factors have been checked for Cronbach’s Alpha. According to Lance, Butts & Michels (2006) it depends on the type of data which Cronbach’s Alpha one should maintain. We maintain a Cronbach’s Alpha of 0.70 due to the fact that the data which is being used is new. However, Lance, Butts & Michels (2006) mention that a Cronbach’s Alpha of 0.7 is not a universal standard of reliability. Due to the huge amount of factors, we have not focused on all of them. The factors have been checked for Cronbach’s Alpha and a correction has been made for the amount of factors; the total amount of factors after testing for reliability is 6.
The questions which have not been used from the survey had a low Cronbach’s Alpha (lower than 0.7). Four questions measuring Tightness in Personnel Controls (Cronbach’s Alpha of 0.608) and three questions measuring Tightness in Results Controls (Cronbach’s Alpha of 0.413) have not been used from the survey. These questions have not been used due to the very low Cronbach’s Alpha. Besides that an attempt has been made to combine the questions with other factors (with a high Cronbach’s Alpha) but this resulted in a low Cronbach’s Alpha (<0.7) of other factors. Thus the decision has been made to exclude these questions from the statistical analysis. The following questions are related to each factor, each factor loading has its own Cronbach’s alpha. This can also be seen in the appendix.
IP_IN_ROLE (Cronbach’s Alpha: 0,905)
Imagine that you are in the role of your supervisor:
o This employee always performs all tasks that are expected of him/her. o This employee always performs all essential duties.
o This employee always fulfills all responsibilities required by his/her job. o This employee always engages in all activities that will directly affect his/her
performance evaluation.
o This employee always meets all formal performance requirements of the job. o This employee always completes all duties specified in his/her job description. o This employee never neglects aspects of the job that he/she is obligated to
23 IP_GENERAL (Cronbach’s Alpha: 0,875)
How would you rate your own job performance in terms of the following: o Quantity of work output
o Quality of work output o Accuracy of work
o Customer service provided (internal and external) o Obtaining personal career goals
o Developing skills needed for your future career o Making progress in your career
o Seeking out career opportunities o Coming up with new ideas o Working to implement new ideas o Finding improved ways to do things o Creating better processes and routines
Tightness in CC (Cronbach’s Alpha: 0,789):
This section looks at your relationship with your organization and your colleagues: o I socialize with my colleagues outside of work.
o Since starting this job, my personal values and those of this organization have become more similar.
o My organization regularly hosts social events for employees. o I am not friends with any of my colleagues.
o I feel a sense of “ownership” for this organization rather than just being an employee.
o My organization communicates its core values to employees. o My organization plans team building events for employees. o My organization creates company sponsored teams for sporting
events/fundraisers/volunteer events. Tightness in PC (Cronbach’s Alpha: 0,755)
This section addresses the hiring process in your organization:
o The hiring process to become employed at my firm is extensive. o You have to go through many steps in order to be hired at this firm.
24 o I interviewed with several people in my organization before being offered a
position.
o The hiring process at my organization evaluates the knowledge, skills, abilities, values and motives of prospective employees.
o Questions left out:
o Before being hired, most of my colleagues and I acquired the same kind of job experience.
o Before being hired, most of my colleagues and I followed the same type of education and training.
o There seems to be little consistency in the type of professional that gets hired for my job.
o The competence of employees within my job title varies greatly. Tightness in BC (Cronbach’s Alpha: 0,711)
These questions address the use of rules and procedures in your organization: o Whatever situation arises, we have existing processes, procedures or rules to
follow in dealing with it.
o Established processes, procedures and rules cover all of my job tasks. o In my organization, we have rules for everything.
o My supervisor frequently monitors the extent to which I follow established process, procedures and rules.
o My job allows me to decide how to adjust rules to best perform my job tasks. o The organization I work in primarily uses established processes, procedures and
rules to give broad guidelines as to how activities are to be performed. o Employees in my organization are encouraged to use procedures flexibly. o Employees in my organization are encouraged to adjust procedures to suit the
situation.
Tightness in RC (Cronbach’s Alpha: 0,772)
These questions address measurement of performance:
o In my job, there is a performance measure for everything.
o My organization sets a large number of performance goals/targets that I am expected to meet.
25 o My supervisor frequently checks to make sure that I am meeting my performance
targets.
o In my organization, employees are expected to meet pre-established goals/targets with no exceptions.
o Approximately how many performance targets are used in the evaluation of your job performance?
o How often do you discuss the results of your performance measures with your supervisor?
o Questions left out:
o In our organization, goals/targets are essentially a guideline rather than a true commitment.
o My supervisor is very considerate of my explanations of deviations from pre-established goals/targets.
o Responding to new, unforeseen opportunities is considered more important by my supervisor than achieving pre-established goals/targets.
IP_IN_ROLE measures the evaluation of individual performance in the role of a supervisor, IP_GENERAL measures individual performance in general, CC measures the Tightness of Cultural Controls, PC measures the Tightness of Personnel Controls, RC measures the Tightness of Results Controls and BC measures the Tightness of Behavior Controls. With these variables we will continue with the statistical analysis. Although both IP_IN_ROLE and IP_GENERAL have a very high Cronbach’s Alpha and they measure individual performance, we chose not to combine these two factors because we assume that there might be different relationships between both variables relative to the rest of the variables. This is based upon the assumption that a person will evaluate him/herself (more) less critically when he/she is evaluating him/herself from (their own viewpoint) the viewpoint of a supervisor.
The control variables (contingency variables) which will be used are size, experience and education. For this research the following questions are used for each control variable. Size is measured by the following question: “How many people are employed by your entire company?” Experience is measured by the following question: “How many years have you worked in your current field?” Education is measured by the following question: “What is the highest level of education you have completed?” A remark has to be made regarding education. Education is an ordinal variable and we approach it as an interval variable.
26
Preparation for hypotheses testing
Table 1 presents the descriptive. Table 1
Descriptive Statistics
N Minimum Maximum Mean Std. Deviation
IP_GENERAL 299 1,92 5,00 3,7144 ,58282 IP_IN_ROLE 299 1,29 5,00 4,1429 ,60508 CC 299 1,25 5,00 3,5503 ,72360 PC 299 1,00 5,00 3,3264 ,82562 BC 299 1,00 4,88 3,0510 ,61449 RC 299 1,00 4,57 2,8117 ,68997 SIZE 299 1,00 4,00 2,8629 1,07656 EXPERIENCE 299 1,00 11,00 6,9331 3,08746 EDUCATION 299 1,00 3,00 1,7358 ,64017 GENDER 299 1,00 2,00 1,3946 ,48959 Valid N (listwise) 299
The amount of people who conducted this survey is 316. We conduct our statistical analysis with a sample of 299, this is the number of respondents who have conducted the whole survey. From this population 39% are female and 61% male. The professionals have an average of almost 7 years of working experience in their field. The average level of education of the professional is a Bachelor’s degree (1.73). The average respondent works in organizations who employ 500 to 5000 people (2.85).
Looking at the mean of IP_GENERAL and IP_IN_ROLE we can see that the means are respectively 3.7 and 4.1. The mean of IP_IN_ROLE (4.1) is higher than the mean of
IP_GENERAL (3.7), this clearly shows that the respondents evaluate themselves less critically from the viewpoint of a supervisor. This is a verification of the assumption that both behave differently.
Table 2 gives an indication of the bivariate relations between the variables of interest. The parametric Pearson and the non-parametric Spearman correlations are presented in the table. Above the diagonal are the non-parametric Spearman correlations and below the diagonal are the parametric Pearson correlations.
27 Table 2 shows that each of the four types of controls and the control variables have different correlations with IP In-Role and IP General. For the parametric correlations we can see that there is a substantial difference. Tightness in CC with IP_GENERAL (0.285) is significant while Tightness in CC with IP_IN_ROLE is not significant (0.076). This is the same for Tightness in PC; correlation with IP_GENERAL is significant (0.228) but not with IP_IN_ROLE (0.106). Another interesting difference is the correlation between Tightness in BC and IP_GENERAL (-0.42) which is negative, and IP_IN_ROLE (0.47) which is positive.
For the non-parametric correlations there is no substantial difference in the correlations between the Control types and IP_GENERAL and IP_IN_ROLE. But, there is a difference in how the Control types correlate with IP_GENERAL and IP_IN_ROLE. We can see that Tightness in RC correlates positively with IP_GENERAL (0.113) but correlates negatively with
IP_IN_ROLE (-0.69).
The above is a verification of the assumption that there is a different relationship between both IP In-Role, IP General and the rest of the variables. Another verification of the assumption is that the correlation between IP_IN_ROLE and IP_GENERAL is not that high for both the parametric (0.378) and non-parametric (0.428) method. Thus we will continue with using individual performance as two variables instead of one.
When comparing both measures regarding which one is better. We have more trust in IP_GENERAL, it has more value compared to IP_IN_ROLE. Firstly, looking at the survey construct of both measures we see that the questions regarding IP_IN_ROLE are more general and the questions regarding IP_GENERAL vary more. This gives a broader type of information. Secondly, the descriptive showed that respondents are more critically when reviewing their own
IP_GENERA
L IP_IN_ROLE CC PC BC RC SIZE EXPERIENCE EDUCATION GENDER
IP_GENERAL ,428** ,281** ,194** -,030 ,113 ,121* ,078 ,046 ,089 IP_IN_ROLE ,378** ,161** ,143* -,002 -,069 ,013 ,111 ,063 ,044 CC ,285** ,076 ,397** -,005 ,159** ,116* -,115* ,047 -,015 PC ,228** ,106 ,402** ,171** ,250** ,185** -,106 ,040 -,095 BC -,042 ,047 ,014 ,151** ,345** ,176** -,084 -,085 ,037 RC ,086 -,062 ,144* ,239** ,359** ,043 -,219** -,141* -,027 SIZE ,136* ,018 ,159** ,192** ,185** ,010 -,024 ,255** -,115* EXPERIENC E ,071 ,137* -,102 -,090 -,079 -,224** -,030 ,018 -,081 EDUCATION ,056 ,026 ,043 ,031 -,070 -,158** ,254** ,042 -,075 GENDER ,066 ,016 -,013 -,100 ,037 -,010 -,126* -,073 -,073
**. Correlation is significant at the 0.01 level (2-tailed). *. Correlation is significant at the 0.05 level (2-tailed).
28 performance (IP_GENERAL) compared to reviewing their performance from the viewpoint of their supervisor (IP_IN_ROLE). Finally, the correlations show that IP_GENERAL has a substantial difference with IP_IN_ROLE; some of the variables correlate significantly with IP_GENERAL and not with IP_IN_ROLE. The statistical analysis will also give more information regarding which measure has more value. Thus, more elaboration on these two measures will be done in the results.
For the control variables we can see that there is a substantial difference between experience, size and individual performance for the parametric variables. For the non-parametric variables there is only a substantial difference between size and individual performance, this has been highlighted in table 2. Besides that, we can see that size correlates positively with CC, PC and BC but not with RC, this is the case for both the parametric and non-parametric correlations. Education and experience correlate negatively with RC, this is the case both the parametric and non-parametric correlations. Experience also correlates negatively with CC, this is only the case for the non-parametric correlations.
According to Field (2014) the central limit theorem applies to large sample data, which implies that the sample mean is approximately normally distributed when the sample data is large (n>30). Besides that the survey is made up of Likert scales, so it is not necessary to check the data for normality. The data has been checked for multicollinearity. One rule of thumb for checking if there is no multicollinearity; is to check whether the correlations between variables should not exceed 0.8, which is the case.
Hypotheses testing
For hypothesis 1 the following variables from the survey will be used: MCS variables, which gives information about the Tightness of the four types of control. Individual performance will also be used. For hypothesis 2 the following variables from the survey will be used: MCS variables, individual performance and gender will be used as the moderating variable. Hypothesis 3 will use the same variables as hypothesis 2.
Hypothesis 1 is tested through a linear regression, the independent variable, Control Tightness, is measured on an ordinal scale. Hypothesis 2 tested through a moderated regression analysis, the independent variable, Control Tightness, is measured on an ordinal scale and the moderator, Gender, is measured on a categorical scale.
For hypothesis 2 the independent and moderating variable will be standardized. After that each independent variable will be multiplied with the moderating variable, this way we can see if
29 gender moderates the relationship between the four types of controls and the individual performance of professionals.
We will test these hypothesis using IBM SPSS. The outputs of SPSS can be found in the appendix. The following paragraphs will have a summation of those outputs.
30 Hypothesis 1
Relation between Tightness in Control Types and Individual Performance
Table 3
H1: IP_GENERAL B t Sig. H1: IP_IN_ROLE B t Sig.
CC 0,173 3,551 0,000 CC 0,050 0,950 0,343 PC 0,097 2,225 0,027 PC 0,082 1,717 0,087 BC -0,106 -1,865 0,063 BC 0,075 1,201 0,231 RC 0,080 1,537 0,125 RC -0,080 -1,396 0,164 SIZE 0,058 1,816 0,070 SIZE -0,012 -0,332 0,740 EXPERIENCE 0,024 2,275 0,024 EXPERIENCE 0,027 2,359 0,019 EDUCATION 0,023 0,432 0,666 EDUCATION 0,012 0,212 0,832 GENDER 0,134 2,026 0,044 GENDER 0,041 0,562 0,575 R² 0,138 R² 0,045 R² adj 0,114 R² adj 0,019 F 5,804 F 1,715 P 0,000 P 0,094
Table 3 gives an overview of the significance of the relationship between the Tightness in the four Control types, the Control variables (IV), General and In-role individual performance (DV) and the Control variables. Besides that it gives a representation of the adjusted R² of the first hypothesis.
We can see that 11.4% of the variance of IP_GENERAL is explained by the four types of controls and the control variables, which is also significant (0.000, p<0.01). Besides that we can see that 1.9% of the variance of IP_IN_ROLE is explained by the four types of controls and the control variables, this effect is marginally significant (0.094, p<0.10).
Table 3 shows that Tightness in Cultural Controls (0.000, p<0.01), Tightness in Personnel Controls (0.043, p<0.05), Experience (0.024, p<0.05) and Gender (0.044, p<0.05) have a significant effect on IP_GENERAL. Besides that Tightness in Behavior Controls is marginally significant (0.063, p<0.10). The other Control types and Control variables don’t have a significant effect on IP_GENERAL. For IP_IN_ROLE we can see that Tightness in Personnel Controls (0.087, p<0.10) is marginally significant and Experience (0.019, p<0.05) is significant.
31 Hypothesis 2
Like mentioned before, the independent and moderating variables have been standardized and multiplied with each other. This had led to four new variables: CCxGENDER, PCxGENDER, BCxGENDER and RCxGENDER.
Moderating effect of GENDER on the relation between Control Types and IP_GENERAL
Table 4
H2: IP_GENERAL B t Sig. H2: IP_IN_ROLE B t Sig.
CC 0,184 3,786 0,000 CC 0,046 0,863 0,389 PC 0,081 1,857 0,064 PC 0,070 1,467 0,143 BC -0,100 -1,764 0,079 BC 0,076 1,225 0,221 RC 0,083 1,575 0,116 RC -0,098 -1,718 0,087 CCxGENDER -0,084 -2,478 0,014 CCxGENDER -0,010 -0,284 0,777 PCxGENDER 0,070 2,022 0,044 PCxGENDER 0,065 1,737 0,083 BCxGENDER -0,030 -0,885 0,377 BCxGENDER 0,032 0,875 0,383 RCxGENDER 0,027 0,779 0,436 RCxGENDER 0,065 1,749 0,081 SIZE 0,057 1,787 0,075 SIZE -0,003 -0,098 0,922 EXPERIENCE 0,025 2,331 0,020 EXPERIENCE 0,027 2,347 0,020 EDUCATION 0,021 0,410 0,682 EDUCATION 0,008 0,148 0,882 GENDER 0,141 2,139 0,033 GENDER 0,047 0,658 0,511 R² 0,162 R² 0,082 R² adj 0,127 R² adj 0,044 F 4,622 F 2,139 P 0,000 P 0,015
Table 4 gives an overview of the significance of the moderating effect on the relationship between the Tightness in the four Control types, the Control variables (IV), General and In-role individual performance (DV) and the Control variables. Besides that it gives a representation of the adjusted R² of the second hypothesis.
12.7% of the variance of IP_GENERAL is explained by the moderating effect on the four types of controls and the control variables, this effect is significant (0.000, p<0.05). For IP_IN_ROLE we see that 4.4% of the variance is explained by the moderating effect on the four types of controls and the control variables, this effect is also significant (0.015, P<0.05).
32 Tightness in Cultural Controls moderated by Gender (0.014, p<0.05) has a significant negative effect on IP_GENERAL. Tightness in Personnel Controls moderated by Gender (0.044,
P<0.05) has a significant positive effect on IP_GENERAL. The effect of Tightness in Personnel Controls moderated by Gender (0.083, p<0.10) on IP_IN_ROLE and Tightness in Results Controls moderated by Gender (0.081, p<0.10) on IP_IN_ROLE are both marginally significant and positive.
Results
The results show that for the first hypothesis there is a significant relationship between Control Tightness and General individual performance (p<0.01) of a professional. There is a marginally significant relationship between Control Tightness and the In-role individual performance (p<0.10) of a professional. Regarding the second hypothesis, there is a significant relationship between Control Tightness and General (p<0.05) and In-role individual performance (p<0.05) of a professional moderated by gender.
The adjusted R² regarding the first hypothesis shows for General individual performance 11.4% and for In-role individual performance 1.9%. For the second hypothesis the adjusted R² shows for General individual performance 12.7% and for In-role individual performance 4.4%. Just like the significance the adjusted R² of In-role individual performance is very low for both hypothesis. Which shows that the results regarding In-role individual performance are not that strong compared to General individual performance.
Both the significance and the adjusted R² give us information regarding which measure is better. In the previous chapters we mentioned that we have more trust in General individual performance compared to In-role individual performance. The outcome of the statistical analysis verifies this. The significance of In-role individual performance is only marginally significant for the first hypothesis (p<0.10). General individual performance is significant for both hypothesis. Besides that, the adjusted R² of In-role individual performance is very low for both hypothesis compared to General individual performance.
33 Table 5
Hypothesis 1 Hypothesis 2
IP_GENERAL IP_IN_ROLE IP_GENERAL IP_IN_ROLE
CC +*** o +*** o PC +** +* +* o BC -* o -* o RC o o o -* CCxGENDER -** o PCxGENDER +** +* BCxGENDER o o RCxGENDER o +* SIZE +* o +* o EXPERIENCE +** +** +** +** EDUCATION o o o o GENDER +** o +** o ***. Significant at 0.01 **. Significant at 0.05 *. Significant at 0.10
Table 5 gives a summation of the significant effects of all the variables on General individual performance and In-role individual performance. This is indicated with a ‘+’, ‘-’ or ‘o’. The first one indicates a (marginally) significant positive effect, the second one indicates a (marginally) significant negative effect and the latter on indicates that there is no significant effect.
In table 5 we can see that regarding the first hypothesis, Tightness in Cultural (p<0.01) and Personnel Controls (p<0.05) have a significant positive effect on General individual performance of a professional. Tightness in Behavior Controls has a marginally significant negative effect on General individual performance (p<0.10).
For the second hypothesis there is a significant positive effect of Tightness in Cultural Controls moderated by Gender (p<0.05) and Personnel Controls moderated by Gender (p<0.05) on General individual performance. In other words, males perform better regarding individual general performance with tighter cultural controls, relative to females. Also, females perform better regarding individual general performance with tighter personnel controls, relative to males. Besides that the positive effect of Tightness in Personnel Controls moderated by Gender (p<0.10) on In-role individual performance and the positive effect of Tightness in Results
34 Controls moderated by Gender (p<0.10) on In-role individual performance are marginally significant. In other words, females perform better regarding in-role individual performance with tighter personnel and results controls, relative to males.
Table 5 shows us that the control variables size is positively significant with General individual performance but not with In-role individual performance. This means that bigger firms demand a higher performance from their employees relative to smaller firms or employees perform better in bigger firms relative to smaller firms.
The control variable Experience (p<0.05), which measures the amount of work experience in the field, has a significant positive effect on both performances. This shows that the more experience a professional has the better his/her performance.
There has not been found a significant effect regarding the relationship between education and the two performances.
The variable Gender also has a significant positive effect on the General individual performance of a professional (p<0.05). The results show that females rate themselves higher regarding general individual performance, relative to males. This is not the case for in-role individual performance. This was not expected because females tend to be more humble. But this shows that females are more confident when rating themselves compared to males.
