The relationship between performance measures and
motivation characteristics
Focusing on financial consultants
Master thesis Accountancy Control, variant Control
University of Amsterdam
Amsterdam Business School
Faculty of Economics and Business
Author: Robert Stolwijk Student ID: 10678689 Version: Final Date of final version: 21/06/2015 Word count: 12812
2
STATEMENT OF ORIGINALITY
This document is written by Robert Stolwijk 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
3
ABSTRACT
A lot of research focuses on the impact of performance indicators on employee motivation. This study is unique and looks at the opposite relation. It looks into the question whether the degree of intrinsic motivation influences the valuation of performance indicators.
This study is the first to combine the self-determination theory and the balanced scorecard. The three components of the self-determination theory, autonomy, competence and relatedness, are used to determine the intrinsic motivation of employees. The performance indicators are categorized in four components consistent with the four perspectives of the balanced scorecard.
Data has been gathered by a survey at a Dutch financial consultancy company. The company has highly educated employees with salaries above average. The survey results from 48 respondents were used to test the hypotheses. This study shows the importance of
experienced relatedness in work. When financial consultants experience relatedness, they value (most) performance indicators more. A likely explanation is that consultants who experience relatedness, feel related to the organization and its clients. These consultants then also support the organization’s goals. The organization’s goals are translated into its performance
indicators, which are therefore also valued by the consultants.
A practical implication of this research is that increasing relatedness will help the financial consultancy firm to get support for its organization’s goals and performance indicators.
The findings of this study are limited by data issues. Further research is needed to confirm the results of this unique study. Most importantly, a bigger number of respondents and a survey held at several companies can check the found relations and find out if results can be generalized.
4
TABLE OF CONTENT
STATEMENT OF ORIGINALITY ... 2
ABSTRACT ... 3
1 INTRODUCTION ... 5
1.1 The effect of performance indicators on employee motivation and results... 5
1.2 The other way around: the effect of motivation on valuation of performance indicators ... 6
2. THEORY AND HYPOTHESES ... 8
2.1 Self-determination theory ... 8
2.2 Balanced scorecard ... 9
2.3 Other relevant variables ... 10
2.4 Hypotheses ... 10
3 METHOD ... 16
3.1 Sample selection ... 16
3.2 Participants ... 17
3.3 Measures ... 17
3.4 Other relevant variables ... 22
4 RESULTS ... 26 4.1 Correlations ... 26 4.2 Regressions ... 28 4.3 Hypotheses - results ... 31 5 DISCUSSION ... 32 5.1 Summary ... 32
5.2 Theoretical implications hypotheses ... 32
5.3 Theoretical implications other significant relations ... 33
5.4 Limitations... 34
5.5 Directions for further research... 36
5.6 Practical implications ... 36
6 REFERENCES ... 38
Appendix A: Survey ... 42
5
1 INTRODUCTION
Motivating employees is one of the most challenging tasks for an employer. Performance indicators play a crucial role in this regard. The relationship between performance indicators and motivation is explained in section 1.1. Less researched is the influence of motivation characteristics on employees’ valuation of performance indicators. Section 1.2 goes into this question. Figure 1 shows these relations schematically.
Figure 1: relations between motivation and performance indicators
1.1 The effect of performance indicators on employee motivation and results
One of the most difficult issues for companies is determining which performance indicators should be used. Although there is a lot of discussion about the effect of assessment systems; most studies find a positive effect of an assessment system on work satisfaction, work motivation and organization involvement (Brown & Benson, 2003; Levy & Williams, 2004; Pettijohn et al., 2001; Tziner & Latham, 1989). Despite the difficulties of performance
measurement tools, employers should therefore give considerable attention to such systems. It can be concluded that an assessment system is one of the most important tools to positively influence the work behavior of employees (Cawley, Keeping & Levy, 1998). Employers find it important to have satisfied employees since a satisfied workforce means better performances, lower staff turnover and less illness among employees (Freeney & Tiernan, 2006; Pettijohn et al., 2001).
Among economists there is a lot of debate on the matter of which performance indicators are best suited to improve employee performance. Most research concerns the balance between financial and non-financial indicators. Challagalla and Shervani (1996) find
Intrinsic motivation
§1.1. Relation, proven in literature and not examined in this thesis
§1.2 Relation, examined in this thesis
Valuation of
6
that the measurement of performances should be based on capacities and behavior. They have shown that assessments based on input, like capacities and behavior have a bigger positive influence on work satisfaction and performances than assessments based on results, like sales and profit. Huang et al (2007) agrees with Challagalla and Shervani (1996) and stated that the output of a company should not only be measured by financial indicators; non-financial indicators should be more important and leading for companies. Huang et al (2007) prove this point by researching the use of non-financial indicators in hotels. They found that there is a relationship between the score on non-financial (customer satisfaction, product quality) indicators and financial indicators. Overall, Huang et al concluded that non-financial measurements are leading for (future) financial results.
Frey & Osterloh, (2012) state that employees will adapt to the incentives given by performance indicators of the organization and neglect other important tasks. When employees neglect other important tasks, the performances of the organization as a whole will decline.
The effect of a performance assessment system is dependent on the view one has on human beings. McGregor (1972) has two different theories: theory X and theory Y. Theory X assumes that people are lazy and only perform when they get a financial return. Theory Y assumes that people like their work, they expect a financial return but also respect for what they do and the possibility to develop themselves. A lot of performance assessment systems are based on the agency theory. The agency theory assumes that the interest of the employee is different than the interest of the company; employees need incentives to work in the interests of the company. Rewards are introduced to enable the employee to do the best for the company (Bonner & Sprinkle, 2002).
1.2 The other way around: the effect of motivation on valuation of performance indicators
The discussion among economists and the conflicting results make it worthwhile to dive deeper into the issue what motivates employees. As mentioned, performance indicators influence motivation. However, it is expected that characteristics of employees also influence their needs and therefore their valuation of performance indicators. For example, Kooij (2010) states that older employees will be less focused on promotion, recognition and salary and more on job security, the usefulness of their work for the society and the possibility to help other people. Knowledge about the impact of employee characteristics (age, motivation, education) on their valuation of performance indicators is very relevant since employees who value performance indicators are more satisfied and will be (once again) better motivated.
7
This thesis looks into the impact of one of the mentioned employee characteristics: motivation. It examines the relation between the valuation that financial consultants attach to different kind of performance indicators and the motivation characteristics of these consultants. More specifically, this thesis looks into the question if employees’ valuation of performance indicators is dependent on the intrinsic motivation of these employees.
The performance indicators are categorized consistent with the principles of the balance scorecard. Research into the link between the motivation of persons and the balanced scorecard perspectives is never done before. This research therefore gives more insight how people value the balanced scorecard perspectives. Results can be used to determine performance indicators which are more in line with the motivation of persons.
Another contribution to the existing literature is the use of the self-determination theory. The self-determination theory is used to determine the intrinsic motivation of the financial consultant. The self-determination theory states that three components are important to be intrinsically motivated, the experienced autonomy, competence and relatedness in work. The self-determination theory is a relatively new theory which is becoming increasingly important in the literature. It is one of the first instances that this theory will be used to determine the motivation and relate this motivation to the valuation of the performance measurement system.
8
2. THEORY AND HYPOTHESES
In this chapter the theory which has been used for the study will be explained. The self-determination theory (Deci & Ryan, 2002) will be used to determine the experienced autonomy, competence and relatedness of employees. The performance indicators will be separated by using the balanced scorecard (Kaplan & Norton, 1992) over the four perspectives of the balanced scorecard: learning & growth, internal, client and financial. Also the control variables used to improve insights will be explained. The second part of this chapter will explain which hypotheses are used and why a positive or negative relation is expected. Figure 2 provides a schematic overview of the relations investigated and the used theories which are further explained in this chapter.
Figure 2: relations investigated and used theories
2.1 Self-determination theory
The first theory which will be explained and discussed is the self-determination theory (Gagné & Deci, 2005). The self-determination theory separates two kinds of motivations: intrinsic and extrinsic. People who are extrinsically motivated only perform when they feel pressure to perform. An intrinsically motivated employee does not need pressure to perform (Gagné & Deci, 2005).
The intrinsic motivation of people can be determined by the self-determination theory. The self-determination theory assumes that people are active organisms who would like to
Valuation of performance indicators (balanced scorecard): -Learning& growth -Client -Internal -Financial Intrinsic motivation following the
self-determination theory: • Autonomy • Competence • Relatedness Relation, examined in this thesis
Used other relevant variables in this study to explain relation: -Intrinsic motivation -Learn & earn
-Extrinsic motivation -Education
-Age -Employed
Relation, but not examined in this thesis
9
learn and develop themselves and integrate new experiences into a coherent sense of self. The social context can support or hinder these desires. Sense-determination theory is focused on the relation between the active organism and the social context. The social context is displayed in three parts: autonomy, competence and relatedness (Deci & Ryan, 2000). Autonomy is the extent to which employees experience a sense of volition and psychological freedom. Competence is the feeling to be effective with your knowledge, skills and attitude. Finally, relatedness means the feeling that employees take care of each other and like each other.
Employees are intrinsically motivated when they experience autonomy, competence and relatedness in work (Deci & Ryan, 2000). Employees who are intrinsically motivated are more engaged in activities, more persistent and more creative. Hence, performances are better (Deci & Ryan, 2000).
Bandura (1997) states that the kind of motivation influences the valuation and therefore the effectiveness of performance indicators. Challagalla and Shervani (1996) state that
performance indicators can have a negative impact on intrinsic motivation when they have a negative effect on the experienced autonomy, competence or relatedness. Therefore, it is expected that intrinsically motivated employees value performance indicators which indicate their achievements on autonomy, competence and relatedness. They value these indicators since they are in line with their own goals.
2.2 Balanced scorecard
The second theory, the balanced scorecard, is used to categorize performance indicators in four categories: learning & growth, internal, client and financial business processes (Kaplan & Norton, 1992). These four perspectives together should represent the strategy of a company. The balanced scorecard is used to quantify the strategy by developing targets for each of the four sections. All the perspectives are related with each other following the theory. For example, a satisfied client of a company is more likely to extend or expand the relationship with the employee / company. This will result in better learning and growth of the employee and the financial situation of the company will improve. Because of the importance of the four perspectives and the interaction between the four perspectives, an assessment cannot involve only one of the four perspectives; learning & growth, internal, client and financial performance indicators should all be part of the assessment.
This thesis looks into the question whether the value a person attaches to performance indicators depends on the level of experienced intrinsic motivation of that person.
10
2.3 Other relevant variables
Other variables (age, education) can also influence the value a person attaches to performance indicators. These indicators are therefore included as variables in the analyses.
First of all, in thesis, intrinsic motivation is determined by the self-determination theory. Another option is to directly ask employees about their level of intrinsic or extrinsic motivation. This was done by asking several questions about this in the survey. This variable was included to get a second view on the impact of motivation characteristics on the valuation of performance indicators.
Besides that, Boxall and Purcell (2008) argue that the kind of work determines the motivation of an employee. An employee can only be intrinsically motivated when the work is connected to the desires of the employees and gives possibilities to learn and growth. In practice employees’ work does not always overlap with the desires of the employee. When work does not overlap with the desires of the employee a motivation change can appear. The variable “employed” is used for the possible motivation change because of the absence of a secondment at a client.
Finally the age of the employee, the years of service at the firm and the education can influence the valuation of the performance indicators. For example, an employee who has just started, has a lower salary and less experience than a colleague with many years of service and might value performance indicators differently (Kooij, 2010). Including the variables age, education and experience in the analyses can give insights in the impact of these variables on the valuation of performance indicators.
2.4 Hypotheses
12 hypotheses will be used to examine the relationship between motivation of financial consultants and their valuation of performance indicators. The hypotheses connect the three components of the self-determination theory to the four perspectives of the balanced scorecard. This leads to 12 hypotheses.
11
Autonomy
Autonomy is the extent to which employees experience a sense of volition, independency, and psychological freedom to decide and to have control over skill utilization (Hackman &
Oldman, 1976). The recognition of volition can be reached by achieving goals that are close to one self.
Different researchers agree that autonomy positively influences learning outcomes. When a person learns a new task, the feeling of personal responsibility will increase which is beneficial for the employee’s motivation to learn (Wright & Cordery, 1999). Furthermore, a high learning motivation can lead to an increased persistence and willingness for the search to new work procedures which can positively influence the learning results of an employee (Nell &
Kozlowski, 2008). So, all in all, autonomy has a positive effect on the willingness to learn and grow (develop) oneself. Therefore it is hypothesized that the experienced autonomy will positively influence the valuation of learn & growth performance indicators (H1a).
The autonomy of employees in an organization can be limited by internal procedures and rules. Because of the internal rules, an employee will experience less sense of choice and psychological freedom when he is carrying out an activity (Deci & Ryan, 2000). This is why is hypothesized that internal performance measures will therefore be valued less by people who experience autonomy (H1b).
A person who has the right to limit the psychological freedom and sense of choice in employee’s work is the client. A client can limit the freedom by a deadline, reward or feedback and can in this way influence the experienced autonomy of the employee. Researchers value the impact of preconditions to the experienced autonomy differently. Taken together,
researchers state that positive feedback neither supports nor limits autonomy (Boggiano & Barret, 1985; Deci, Cascio & Krusell, 1975). However, a client will support the autonomy of the employee when the client assumes that the employee has the same goal as himself. Several studies tell that the employee will experience a higher sense of autonomy when this is the case (Grolnick & Ryan, 1989). So, it can be hypothesized that employees who experience autonomy have clients with the same goals as they have. This is why it is hypothesized that people who experience autonomy will value client performance indicators (H1c).
When an employee is intrinsically motivated, which means a (high) experience of autonomy, competence and relatedness (self-determination theory), financial indicators can lower the intrinsic motivation of the employee (Gneezy et al., 2011). This is because financial performance indicators can be seen as control mechanisms (Prendergast, 2008). As mentioned before control leads to a decreased sense of volition and experienced psychological freedom of
12
the employee. This is why it is hypothesized that people who value autonomy will not value financial performance indicators (H1d).
This leads to four hypotheses which link the autonomy of the self-determination theory to the four perspectives of the balanced scorecard.
Hypothesis 1a: The more financial consultants experience autonomy, the more they value performance indicators measuring learning & growth (positive relationship).
Hypothesis 1b: The more financial consultants feel autonomy, the less they value performance indicators measuring internal functioning (negative relationship).
Hypothesis 1c: The more financial consultants feel autonomy, the more they value performance indicators measuring client satisfaction (positive relationship).
Hypothesis 1d: The more financial consultants feel autonomy, the less they value performance indicators measuring financial results (negative relationship).
Competence
Competence is the feeling to be effective; using your knowledge, skills and attitude. Competence also encompasses the desire to extend the environment and to engage in
challenging tasks to test and to extend skills (Deci & Ryan, 2000; White, 1959). Learning & growth performance indicators normally measure whether employees have extended their environment, have done challenging tasks and have extended their skills. In short, people who experience competence will value learning & growth performance indicators (H2a).
Roth (1997) shows that the extent to which an organization is focused on competence is dependent on internal characteristics. Examples of internal characteristics are communication, management operating style and management philosophy. These internal characteristics can be controlled and measured by internal performance indicators. It is expected that competent employees are most effective/satisfied in a well-functioning organization. That is why it will be hypothesized that people who experience competence will value internal performance
indicators positively (H2b).
The competences of employees can be best assessed by clients since they are in the position to benchmark performances of different providers of products and services (Bernardin et al, 1998). Positive feedback of other people, like clients, is positively related to the
experienced competence of employees. This is the case because positive feedback satisfies the inborn need for competence (Mouratidis, Vansteenkiste, Lens, & Sideridis, 2008). Negative feedback works also positive to improve competence, provided that the negative feedback is
13
focused on change. Negative feedback which is focused on change is positively related to intrinsic motivation (Carpentier and Mageau, 2013). Hypothesized in this study is that people who experience competence will value feedback and will therefore value client performance indicators positively (H2c).
Rewards based on competences increase the intrinsic motivation of employees (Rosenfield, Folger & Adelman, 1980). A higher pay leads to greater willingness to work on the same task in the future. Rewards which are not based on competence lead to a lower intrinsic motivation state researchers. It is expected that financial performance indicators are generally not based on competence. This is because financial performance indicators are not directly linked to self-development (increased competence). Therefore financial indicators generally don’t support the intrinsic motivation of employees. This is why a negative relation is expected between experienced competence and financial performance indicators. (H2d).
Hypothesis 2a: The more financial consultants experience competence, the more they value performance indicators measuring learning & growth (positive relationship).
Hypothesis 2b: The more financial consultants feel competence, the more they value performance indicators measuring internal functioning (positive relationship). Hypothesis 2c: The more financial consultants feel competence, the more they value performance indicators measuring client satisfaction (positive relationship).
Hypothesis 2d: The more financial consultants feel competence, the less they value performance indicators measuring financial results (negative relationship).
Relatedness
Relatedness means the feeling that employees take care of each other and like each other. People experience relatedness when they develop a sense of communion and intimate relationships at a social context (Deci & Ryan, 2000). An example of a social context is the work environment. It is assumed that people would like to integrate in a social context to benefit from common desires and goals (Attachment Theory, Bowlby 1969). People who experience relatedness at work feel connected with the organization and likely feel connected to the goals of this organization.
An usual goal of an organization is to be efficient and effective with all the resources, like the employees of the organization. To keep the organization efficient and effective, learning& growth of employees is important. The learning & growth component is important because, the last years, the economy has become a knowledge economy (Kessels, 2001). It has
14
become more important to update the knowledge of employees and to use this knowledge effective and efficient (Cluitmans, 2002). Learning and growth performance indicators can contribute to an effective and efficient organization. Financial consultants who value relatedness also value the organization’s goals and therefore appreciate learn & growth performance indicators positively (H3a).
These resources can only be used effective and efficient if the organization
continuously keeps changing. Daly et al (2003) state that the internal organization is important for changing organizations. Internal performance indicators can contribute to the
well-functioning of an organization. It is hypothesized that financial consultants who value relatedness therefore also value internal performance indicators positively (H3b).
An organization needs to learn and grow and use resources effectively to perform well and to attract and to satisfy clients. The clients are best suited to value the performance of the organization and its employees (Bernardin et al, 1998). The reason clients are best suited to assess employees is because they are in the position to benchmark performances of different providers of products and services. This holds especially in the service sector since the service sector is a high knowledge sector. In such a sector, there is a direct relationship between the performance of an employee and the satisfaction of a client (Cardy & Dobbins, 1994).
Therefore a good client rating is one of the organization’s goals; feedback of the client will be valued by the employee. This is why is hypothesized that people who experience relatedness will value client performance indicators (H3c).
The work organization where people would like to integrate in and to benefit from cannot prosper without good financial results. To ensure a financially healthy company
financial performance indicators will be used. Because of the relatedness with the organization the employee will value these financial performance measures (H3d).
To conclude, it can be said that employees who feel related to the organization will attach more value to performance measures intended to keep the organization healthy. The trust in these performance measurements will be high, and because of the high involvement better scores are expected. This leads to four hypotheses which link relatedness of the
self-determination theory to the four perspectives of the balanced scorecard.
Hypothesis 3a: The more financial consultants experience relatedness, the more they value performance indicators measuring learning & growth (positive relationship).
Hypothesis 3b: The more financial consultants feel relatedness, the more they value performance indicators measuring internal functioning (positive relationship).
15
Hypothesis 3c: The more financial consultants feel relatedness, the more they value performance indicators measuring client satisfaction (positive relationship).
Hypothesis 3d: The more financial consultants feel relatedness, the more they value performance indicators measuring financial results (positive relationship).
16
3 METHOD
To answer the hypotheses a survey was conducted at a Dutch financial consultancy firm. The first part of this chapter gives a description of the company and tells why this company is considered a good fit for this research. The second and third part of this chapter are about the characteristics of the collected responses and the used constructs to determine the motivation of the consultants and their valuation of the different performance indicators. Finally, the used constructs for other relevant variables included in the analyses will be explained.
3.1 Sample selection
The sample is collected at a Dutch financial consultancy firm. The firm was founded in 1998 in The Hague. The firm is a financial secondment agency with 56 university level financial
consultants working for clients. The services, which are provided by the financial consultants, are the deployment of highly educated personnel in order to provide support for external organizations. These services comprise of advice on several hierarchical levels. The clients are both profit and non-profit organizations. Examples are financial institutions, local authorities, health institutions and central government. The turnover is approximately 6,5 million euro in 2014.
A survey is used to find out how the 56 consultants of the financial consultancy
company are motivated and which performance indicators they value most. As mentioned, the financial consultants are university educated. The high education level is important for this study. After all, literature tells that highly educated employees consider developing themselves as one of the most important work conditions (Van Hoof, 2002). Secondly, since highly
educated people earnings’ are above average, they are less extrinsically motivated. It is therefore expected that the respondents of this survey will be more intrinsically than extrinsically motivated.
There are four functions at the financial consultancy company used in this study: junior consultant, medior consultant, senior consultant and the learn and earn function. The
experience of the consultant determines the function: junior, medior or senior. The fourth function is the learn and earn function. A consultant can choose for the learn and earn function. A learn and earner is only assessed on financial performance/ turnover and he or she receives a standard percentage of his own turnover. This means that, in this regime, a consultant who works a lot of hours receives more salary. However, no deployment at a client means no extra salary (a minimum salary is agreed). The regime for other financial consultants is different:
17
they are assessed on multiple performance indicators and can get a salary increase if they perform well on these indicators. The difference in assessment methods between a learn and earner and all other financial consultants makes it possible to see if learn and earners value the balanced scorecard significantly different than all other consultants (who get assessed on several kinds of indicators).
3.2 Participants
The survey has been sent out to all the 56 financial consultants of the company by email. They were given two weeks to respond. One week after the initial invitation a reminder was sent. One day before the deadline, the consultants were called to remember / ask them to fill in the survey.
To make sure that consultants would answer the questions in the survey honestly, two promises were made. First, it was promised that the data gathered by the survey would not be reported on an individual level; only means would be analyzed. Besides that, it was mentioned that only the total results of the entered data would be shared with the board of the company. The exact purpose of the survey was left out to prevent employees being biased when filling in the survey. This was done by only indicating in the mail that the survey is about the work perception of financial consultants.
To avoid confusion, the survey clearly mentioned that the questions are specifically applicable to the work environment of the consultancy firm; not to the work environment of the client. At the deadline date there were 48 responses, which is a response percentage of 86 percent.
3.3 Measures
Several ways of asking questions were used in the survey. Apart from the general questions like function, education level and age, all the questions needed to be answered on a 7 point Likert scale (Likert, 1932). All questions and answers were stated in Dutch. The results of the survey were analyzed by using SPSS.
Self-determination theory
To measure the autonomy, competence and relatedness, the self-determination scale of Deci and Ryan (2002) was used. This scale consists of 21 statements. These statements determine the score of an employee/consultant on the three components of the theory. Out of the 21
18
questions, nine are reverse scored. Appendix A displays which (reversed) statements are questioned.
The first step in analyzing the data was the principal component analysis (Table 3). All statements together loaded on seven different components. Factor analyses for each component of the self-determination theory was conducted to find out which statements did not load on their own components (autonomy, competence and relatedness). These statements were deleted. After that, all the statements together still loaded on more than three components. Different factor analyses and correlation analyses have been conducted to find out which statements needed to be deleted. Finally, after deleting eight statements, the statements still loaded on five different components (Table 3). The remaining five statements of competence did load on three different components. Deleting more than eight statements, however, did not solve this problem or lead to improvement of the cronbach alpha. Therefore no more
statements were deleted. In appendix A is displayed which statements were deleted. The total explained variance of the remaining 13 statements is 66%. In Table 1 a summary of the reliability of the data is given.
Table 1: reliability by component Number of
statements
Explained variance % Cronbach Alpha
Autonomy 4 45,35 0,56
Competence 5 56,60 0,46
Relatedness 4 51,20 0,66
A rule of thumb is that the cronbach’s alpha should be at least 0,5 and preferably above 0,7 to have a reliable measure. The relatively low explained variance and cronbach alpha of Table 1 necessitates a robustness check to confirm that the right statements have been used. This robustness check consists of three statements. Statement 45 is used for the component autonomy, statement 40 for competence and statement 37 for relatedness. These statements are chosen since they display the experienced autonomy, competence and relatedness of the consultant as best as possible; these statements are most in line with the theory of the three components. The robustness check shows some different (significant) correlations and (significant) regressions than the correlations and regressions of the version that uses more statements to determine the components: autonomy, competence and relatedness. Appendix B
19
displays the correlation and the regression results of the robustness check with the three used statements.
An explanation for the different results of the robustness check can be the various work locations and work environments of the consultants. In the survey, it is mentioned that the questions are specifically applicable to the work environment of the consultancy firm.
However, a statement like, “I pretty much keep to myself when I am at work”, might be filled in by the consultant in the context of the clients’ work environment. This is quite logical when a consultant is employed at the client for a long time. Especially, the robustness check is sensitive to this effect because the robustness check consists of only one question per component. It is expected that the influence of different work environments will be less by using more statements per component since not all statements are equally sensitive to the influence of different work contexts. For example the statement “I have been able to learn interesting new skills on my job”, is dependent on the client’s work context but also dependent on the possibilities the consultancy firm creates for the consultant.
Since the negative impact of various work locations on the results is expected to be less when using the variant with 13 statements, this variant is chosen. However, the negative impact of different work contexts on results will not completely be eliminated. This limitation will also be mentioned in chapter five. In that chapter it will also be mentioned how future research can take different work contexts into account.
Balanced Scorecard
The statements for the balanced scorecard are derived from the assessment system of the financial consultancy company used in this study. Since it is expected that financial consultants are intrinsically motivated, statements which are related to the enhancement of autonomy, competence and relatedness should be valued more by them than other statements. After all, theory tells that an intrinsically motivated employee should like it when they get assessed on personal characteristics and results which can contribute to the experienced autonomy, competence and relatedness.
The statements cannot be clearly allocated to the four perspectives of the balanced scorecard (client, financial, learning & growth and internal). Therefore, the statements did not load correctly on the four components (principal component analysis, Table 3). For example, the communication between the consultancy firm and the consultant can be seen as an internal perspective and as a client perspective (the financial consultancy firm is a kind of client for the consultant). Four statements were removed to make sure the statements loaded on the four
20
perspectives of the balanced scorecard. The removed statements are: statement 14 (learning & growth), 20, 22 and 26 (internal). The exact statements are shown in Appendix A (survey). The total explained variance of the remaining nine statements is 83,22%. A summary of the
reliability of the data by component is given below. In this case cronbach’s alpha indicates that the measures are reliable.
Table 2: reliability by component Number of statements
Explained variance in % Cronbach Alpha
Financial 2 83,91 0,81
Client 1 Inapplicable Inapplicable
Learning & Growth 2 75,66 0,67
21
Table 3: principal component analysis
Α N M SD Min Max Rotated component matrix (absolute valued > 0.4) 1 2 3 4 5 Autonomy .56 49 4.84 0.91 V29 48 5.40 0.98 3 7 0.71 V33 49 4.92 1.79 1 7 0.67 V39 49 3.53 1.40 2 7 0.52 V48 49 5.55 1,28 1 7 0.72 Competence .46 49 5.24 0.69 V31 49 5.57 1.44 1 7 0.89 V32 48 5.38 1.12 1 7 0.48 V38 48 5.29 1.25 1 7 0.86 V40 48 5.00 0.97 2 7 0.68 V42 49 4.96 1.34 1 7 0.88 Relatedness .66 48 5.15 0.69 V30 48 4.50 1.22 1 7 0.74 V37 48 3.46 1.52 1 7 0.84 V43 48 4.50 1.22 2 7 0.61 V49 48 5.83 0.66 4 7 0.48
Own value criterion; Rotated on 4 components
Learning & Growth .67 48 4.99 1.20 V10 48 4.83 1.49 2 7 0.93 V12 48 5.15 1.25 3 7 0.60 Internal .68 48 5.13 1.27 V18 48 4.92 1.61 1 7 0.86 V24 48 5.33 1.29 3 7 0.75
22 Client V8 48 6.13 1.45 5 7 0.94 Financial .81 48 5.31 1.31 V16 48 5.54 1.47 1 7 0.85 V28 48 5.08 1.38 1 7 0.89
3.4 Other relevant variables
The other relevant variables used in this study are motivation (intrinsic and extrinsic), the number of months employed at a client during last year, age, years of service, function and education. These variables were introduced in section 2.3 and were added because they might help explain unexpected results and give alternative explanations or insights. This paragraph will show how these variables were measured and what the results for these other variables are. Questions have also been asked about the opinion of the team manager (Appendix A). In the end, however, this indicator has not been included in this thesis because the results don’t contribute to answering the research questions of this study. Besides that, the limited number of respondent’s necessitates making choices in (the number of) used variables for the
regression analysis.
Motivation
12 statements where used to determine whether financial consultants are intrinsically or extrinsically motivated. These 12 statements were originally developed by Gagné et al (2010). The statements load on three different components. However, they should load on two
components: intrinsic and extrinsic motivation. The problem seemed to originate from the extrinsic part; the six statements which were used to measure the extrinsic motivation did not load on a single factor. That’s why two statements used to determine extrinsic motivation were removed. After removing two statements the explained variance is 67% overall. The explained variance for the extrinsic measure is 68% and the measure has a cronbach alpha of 0,84. In the response for this study all the intrinsic motivations statements load on a single factor wherein 67% of the variance is explained and the cronbach alpha is 0,88. In appendix A the two deleted statements are displayed. Table 4 shows how financial consultants score on intrinsic and extrinsic motivation. The difference between the scores on intrinsic (mean 4.86) and extrinsic
23
motivation (mean 3.07) is significant (p < .05). As expected, financial consultants of the firm are mainly intrinsically motivated.
Table 4: principal component analysis of control variables
Α N M SD Min Max Rotated component matrix (absolute valued > 0.4) 1 2 Intrinsic motivation .88 48 4.86 .98 V50 48 4.42 1.23 1 7 .81 V51 48 5.04 1.30 1 7 .82 V52 48 4.38 1.52 1 7 .77 V53 48 4.98 1.08 3 7 .72 V54 48 5.17 1.12 2 7 .77 V55 48 5.21 1.18 2 7 .77 Extrinsic motivation .84 48 3.07 1.26 V56 48 2.94 1.49 1 6 .83 V57 48 3.31 1.49 1 6 .85 V58 48 3.54 1.66 1 7 .80 V59 48 2.48 1.44 1 6 .75
Function and learn & earn
As mentioned, four kinds of functions are possible: junior consultant, medior consultant, senior consultant and the learn and earn function. The assessment of consultants in the learn and earn function is different than the assessment of consultants in the other three functions. The
function of the financial consultant is used as control variable in two ways. The first variable is called “function”. This is a variable which is one for junior, two for medior, three for senior, the learn and earner is listed as missing in this variable. The second variable distinguishes learn and earn from junior, medior and senior consultants together. This is a dummy variable called “learn & earn” which is one for: junior, medior and senior consultants and zero for the learn and earners.
24
These variables have been included to find out what the impact of different assessment methods is and to control for this difference.
Number of months seconded
This control variable gives the number of months the financial consultant was seconded at clients during the last year.
Age, years of service and education
The variables age and years of service are both questioned in number of years. Age and years of service are correlated since they both measure more or less the same. The answers of these questions differ very much between consultants. To make the answers more valuable, the answers have been recoded in groups. These groups were formed in such a way that all four groups are of almost equal size. Different methods have been conducted to find out whether the variable age or the variable years of service should be used. In the end, it was found that using the variable age, divided in four groups, was most insightful. Therefore, only the variable age will be used in the analyses. The four used age groups are displayed in Table 5.
Education is included to determine if there is a difference in motivation and value given to the performance measures between consultants with a bachelor and a master degree.
25
Table 5: Characteristics respondents Mean SD Min Max
Function¹ Junior Consultant 8 2.07 0.68 1 4
Medior Consultant 23
Senior Consultant 11
Learn & Earn 6
Learn & earn² Consultant 42 1.13 0.33 0 1
Learn & earn 6
Education level³ Bachelor degree 31 1.35 0.48 1 2
Master degree 17
Age4 24 - 29 10 2.58 1.09 1 4
30 - 32 12 33 - 38 14 39 - 49 12
Number of months seconded at a client last year 9.71 2.9 3 12
¹ Function is scored 1 for junior, 2 for medior, 3 for senior and 4 for learn& earn ² Consultant (junior, medior, senior) is scored 1, learn & earn is scored 2
³ Education is scored 1 for Bachelor, 2 for Master
26
4 RESULTS
The chapter starts with correlation analysis. The second paragraph gives the results of the regression analyses. The third and final paragraph provides an overview of each hypothesis of this study and whether it is rejected or not rejected.
4.1 Correlations
The Spearman and Pearson correlations between independent variables is shown in table 7. First of all, the correlation matrix indicates if collinearity between independent variables exist. Collinearity needs to be avoided since it influences the beta coefficients. A sign of collinearity is a correlation that exceeds 0.5. Luckily, the matrix does only show two early signs of
collinearity. The correlation between the independent variable financial and the independent variable learn& growth exceeds a correlation of 0.5. However, this correlation is not very strong because it is only the case in the non-parametric Spearman’s correlation and is therefore not considered harmful. The second exception is the correlation between control variable age and control variable function: in both methods their correlation score is .76. This means that there is a coherence between the variable function and the variable age. This coherence is expected since the variable function measures partly the same as the learn and earn variable (see “function” in section 3.4). To avoid collinearity and to limit the number of independent variables, the variable function is omitted from the regression analyses.
Secondly, the correlation matrix gives a first impression of the relations between the balanced scorecard and the self-determination theory. Table 6 summarizes the significant correlations between the components of the balanced scorecard and the components of the self-determination theory.
Table 6: Summary of significant correlations
***P<.01; ** P<.05; *P<.10; p-values are two tailed. † P<0.10; p-values are one tailed
Learning &
growth Internal Client Financial
Autonomy - Pearson† -
Competence - - Spearmann† -
Relatedness - Pearson*** and
Spearmann**
Pearson** and
27
Table 7: Variable correlations
*** P<.01; ** P<.05; *P<.10; p-values are two tailed. † P<0.10; p-values are one tailed
28
4.2 Regressions
Correlation, however does not imply causality, since it does not control for other variables. Regression analysis is used to determine possible dependencies. That is also the reason why, subsequently, linear regression analysis was conducted (Table 9). Different kind of regression analyses have been conducted to find significant relations and to check the robustness of these relations. All regressions use one of the four components of the balanced scorecard as the dependent variable. The first model shows the results of the regression with only the three components of the self-determination theory as independent variables. The second model is a regression with the self-determination theory components and the variables intrinsic and extrinsic motivation. The third model shows the results of the regression with the
self-determination components and all variables together. The fourth model is the same regression without self-determination theory variables.
The results are based on model 1 (regressions of Table 9). Model 1 is used because of the limited number of independent variables in this model. This limited number was desired since a higher number of independent variables means a lower adjusted R square. Low adjusted R squares are possible when the number of independent variables is relatively high compared to the number of respondents. Theory advises to limit the number of independent variables when there is a limited number of participants. However, a specific ratio between respondents and independent variables is not clearly given by theory. A low adjusted R square shows that the variables are hardly interrelated; the findings therefore have less explanatory power. A lower number than three independent variables was not possible because of the used self-determination theory. The self-self-determination theory consists of three components which means a minimum of three independent variables as included in model 1.
Using model 1 means that no other variables are part of the regression, as was
originally intended. However, this is not considered a problem. Coefficients and significance of coefficients for the three investigated independent variables (autonomy, competence and
relatedness) hardly change when switching from model 1 to model 2 or model 3, which do include other variables. Still, when the dependent variable is also significantly related to the independent variable in models 2 and 3, this is additional proof of a significant relationship.
As an extra, model 4 is used to determine the relations between other variables and the dependent variables. All significant relationships are highlighted below.
29
Significant relations
The hypothesized positive relation between the experienced relatedness and the valuation of learning& growth performance indicators is only marginally significant (β = .19; p < 0.20) in model 3 and is not significant in models 1 and 2. The hypothesis is therefore rejected.
People who experience relatedness value internal performance indicators significantly more (β = .36; p < 0.05). This positive relation is also significant in model 2 (β = .31; p < 0.05) and even more significant in model 3 (β = .42; p < 0.01).
The linear regression of model 1 supports the hypothesized positive effect of relatedness on the valuation of client performance indicators (β = .32; p < 0.05). This relationship is confirmed by model 2 (β = .32; p < 0.1) and model 3 (β = .32; p < 0.1).
People who experience relatedness value financial performance indicators more (β = .34; p < 0.05). This support is still significant in model 2 (β = .31; p < 0.1) and model 3 (β = .34; p < 0.1).
The hypothesized positive relation between the valuation of learning& growth
performance indicators and autonomy is rejected in models 1 and 3 and is marginally negative (instead of hypothesized positive) in model 2 (β = .31; p < 0.1).
Finally, the hypothesized positive relation between the experienced autonomy and the valuation of learning & growth performance indicators is only marginally significant (β = -.17; p < 0.20) in model 2 and is not significant in models 1 and 3. The hypothesis is therefore rejected.
Table 8: summary significant relations according to regression analyses
Learning &
growth Internal Client Financial
Autonomy 2† - - -
Competence 3† - - -
Relatedness 3† 1**, 2**, 3*** 1**, 2*, 3* 1**, 2*, 3*
Although, no hypothesizes have been made for the other variables, several significant relations have been found and are, as a bonus, shortly mentioned. People with a high extrinsic motivation value internal performance indicators less according to model 2 and model 4 (see Table 9). The second significant relation is that learn& earn consultants value internal performance indicators (model 3 and model 4) and learning& growth performance indicator (model 3 and model 4) significantly less than other consultants (see Table 9). The same
30
growth performance indicators (model 3 and model 4) and internal performance indicators (model 3). Finally, the age of the consultant is marginally positive related to the valuation of internal performance measures in model 3; older consultants’ value internal performance indicators more than their younger colleagues in that model (see Table 9).
31
4.3 Hypotheses - results
Table 10 gives an overview of the hypotheses of this study and whether they are rejected or not rejected.
Table 10 Hypotheses
***p<.01; **p<.05; *p<.1 two-tailed; † p<0.1 one-tailed. (based on Table 9, model 1).
Hypothesis Result
1a The more financial consultants experience autonomy, the more they value performance indicators measuring learning & growth (positive relationship).
Rejected
1b The more financial consultants feel autonomy, the less they value performance indicators measuring internal functioning (negative relationship).
Rejected
1c The more financial consultants feel autonomy, the more they value performance indicators measuring client satisfaction (positive relationship).
Rejected
1d The more financial consultants feel autonomy, the less they value performance indicators measuring financial results (negative relationship).
Rejected
2a The more financial consultants experience competence, the more they value performance indicators measuring learning & growth (positive relationship).
Rejected
2b The more financial consultants feel competence, the more they value
performance indicators measuring internal functioning (positive relationship).
Rejected
2c The more financial consultants feel competence, the more they value performance indicators measuring client satisfaction (positive relationship).
Rejected
2d The more financial consultants feel competence, the less they value performance indicators measuring financial results (negative relationship).
Rejected
3a The more financial consultants experience relatedness, the more they value performance indicators measuring learning & growth (positive relationship).
Rejected
3b The more financial consultants feel relatedness, the more they value
performance indicators measuring internal functioning (positive relationship).
Not
Rejected** 3c The more financial consultants feel relatedness, the more they value
performance indicators measuring client satisfaction (positive relationship).
Not
Rejected** 3d The more financial consultants feel relatedness, the more they value
performance indicators measuring financial results (positive relationship).
Not
32
5 DISCUSSION
This chapter discusses the results of this study. The first section is a summary of the study followed by the theoretical implications, limitations, directions for further research and the practical implications.
5.1 Summary
This section contains a summary of the findings. 12 hypotheses were made. These hypotheses connected the four components of the balanced scorecard with the three components of the self-determination theory. Three hypotheses were proven by regression analyses. They all involve relatedness. The study found significant positive relations between relatedness and internal performance indicators, client performance indicators and financial performance indicators. In short, the more financial consultants value relatedness the more they value these performance indicators. The hypothesized positive relation between the experienced
relatedness and learning& growth performance indicators was not found to be significant. All hypotheses involving autonomy and competence were not supported by the analyses.
5.2 Theoretical implications hypotheses
Theory tells that personal characteristics influence the valuation of performance indicators. This study shows that the experienced relatedness is one of those personal characteristics. However, autonomy and competence do not have a significant impact on the valuation of performance indicators according to this study.
The importance of relatedness is remarkable. Financial consultants who experience relatedness value internal, client and financial performance indicators significantly more. As mentioned before relatedness is the feeling that employees take care of each other and like each other. An explanation for the importance of relatedness compared to competence and
autonomy is not given by theory. An explanation might be that a financial consultant considers clients and the internal organization as relations. A consultant feeling related would want to satisfy its relations: the client and his/her employer. Performance indicators measuring client satisfaction and the functioning of the internal organization and financial results are therefore important to the employee. A possible explanation for the finding that experienced relatedness is not linked to learning & growth performance indicators is that relatedness involves other persons (relations) while learning & growth performance characteristics measure individual achievements.
33
The lack of significance between autonomy and the components of the balanced scorecard is possibly due to the fact that autonomy is not really captured in the assessment of the financial consultant company. Therefore, it is hard to find a logic relation between autonomy and the balanced scorecard.
At last, the absence of a significant relationship between competence and the components of the balanced scorecard might be caused by the specific situation of the consultants in this survey. They work for different clients over time and the feeling of competence may differ depending on the client. In the survey it was clearly mentioned that consultants should fill in the survey based on the work environment of the consultancy firm. However, when a consultant is seconded at a client for a long time, he/she might fill in the survey based on the client’s work environment (see also section 3.3). If a consultant works at a client who does not give him a feeling of competence this leads to a lower score on
competence. However, the consultant can still feel competent and be highly intrinsically motivated. After all, the consultant knows that the current situation (which he/she probably has filled in in the survey) is only temporary and therefore not representative.
5.3 Theoretical implications other significant relations
In this study several relations were investigated that were not part of the main focus of this study and were not captured in hypotheses. These relations are now discussed.
The regression analyses of this study shows that learn and earners value learning& growth and internal performance measures significantly less than other consultants. Other consultants may value these components more because they get assessed on these indicators while learn and earners are not assessed on these indicators. Learning & growth and internal performance measures are therefore not important to learn and earners. These results are in line with the findings of Frey & Osterloh (2012). They state that employees will adapt to the
incentive system and neglect other important tasks. As mentioned, this conclusion can also be drawn in this study: learn and earners who get only assessed on financial incentives care less about the internal organization and their own learning& growth.
The longer a consultant is seconded at a client, the less importance he/she attaches to internal and learning& growth measures. After all, a consultant who is seconded has less (reason for) communication and connection with his/her own consultancy organization. Therefore, the importance of the internal organization / progress will decrease when the consultant is longer seconded.
34
The significant negative relation between months seconded in a year and valuation of learning & growth performance indicators could be interpreted as follows. When a consultant starts at a client, he/she will learn new tasks / new people because of the new environment. Therefore, when a consultant is seconded he/she learns and grows automatically (on the job). The learning & growth curve will level off when he/she is longer seconded; the new tasks and people will become routine for the consultant. Boxel and Purcell (2008) state that an employee can only be intrinsically motivated when the function is connected to the desires of the
employees and gives possibilities to learn and grow. When a consultant is seconded more months, the desires of the employee to learn and grow will level off and he/she will value learn & growth performance indicators less. This explains the negative relation of learning& growth performance indicators with the months of secondment.
When a consultant is not seconded, there is no possibility to learn and grow. The absence of learning and grow possibilities is missed most by these (high educated) consultants and therefore they will value indicators related to learning and growth more.
The older consultants value internal performance measures more than their younger colleagues. Kooij (2010) stated that the goals of people are dependent on their age. For
example, older employees value job security more than younger colleagues. An explanation for the higher value older people attach to internal performance measures can be that younger people change jobs more often. Therefore the relationship of young employees with the organization is less than the relationship of older employees with the organization. Besides that, older employees value job security more which makes the survival of the company more important; internal performance measures can enable survival.
5.4 Limitations
The first and most important limitation is the limited number of respondents. The limited number of respondents is due to the limited number of financial consultants at the company used in this survey. After all, of the 56 consultants the survey was sent to, 48 responses were received. This means a very good response rate of 86%. The high response rate also implies there is no reason to assume outcomes would be different when everyone would have filled in the survey. The limited number of respondents is an important explanation for the low
Adjusted R Square. This raises questions about the data quality. The low number of respondents also was a reason to limit the number of independent variables used in the regressions. Therefore, model 1 was used for testing the hypotheses, like explained in section
35
4.2. A drawback of model 1 is that no other (control) variables are included to further explain the relation between the components of the self-determination theory and the components of the balanced scorecard.
The second limitation is that the data to determine the experienced autonomy, competence and relatedness of the financial consultants gave some issues, as explained in section 3.3. Research into the link between the motivation of employees determined by the self-determination theory and the balanced scorecard perspectives is never done before. This study used the self-determination theory to determine the experienced autonomy, competence and relatedness of employees. 21 statements were used to determine these three components, nine items were reverse coded. These reversed statements should act as speed bumps and control questions (Podsakoff, MacKenzie, Lee, & Podsakoff, 2003). The effect should be that the respondent slows down and thinks about the question and answer. However, it can also happen that some respondents missed the negative wording and created artificial responses and therefore caused some method bias. Another reason why the data gave some issues can be that the employees of the financial consultancy company are working for clients. In the survey is clearly stated that the questions are applying on the consultancy firm, not the client. However, it is difficult to distinguish the work environment of the consultancy firm to the work
environment of the client. This can create divergence in responses and therefore also cause some method bias.
The third limitation concerns the balanced scorecard. The balanced scorecard consists of four components: learning & growth, internal, client and financial. Dividing the
performance indicators of this company over the four components of the balanced scorecard has never been done before. This method seems to have worked since the performance
indicators of this company were, after deleting some statements, successfully divided over the four components of the balanced scorecard. Still, the lack of previous experience with this method makes it hard to tell if all considerations have been taken into account.
The fourth and last limitation concerns the measure of the control variable
intrinsic/extrinsic motivation. The principal component analysis showed that two of the six statements used to determine the extrinsic motivation loaded on another component. These two statements were dropped from the analysis, which raises questions about the validity of the measurement method of intrinsic/extrinsic motivation.
36
5.5 Directions for further research
The combination of the self-determination theory and the components of the balanced
scorecard is made for the first time. Because it is done for the first time, improvements can be made. First of all, a bigger sample and a survey conducted at multiple firms is needed because of several reasons. A bigger number of respondents makes it possible to overcome the data issues in this study. Besides that, a bigger sample and more firms are needed to generalize the results of this research. Further, conducting the survey at companies not involved in seconding employees prevents problems with different work environments of employees . Lastly,
including more companies in one study will also make it possible to use differences between companies to determine and explain differences in motivation between employees. These differences can determine critical factors that influence the motivation of employees.
Other interesting questions for further research are: What happens with the results when the company is not a financial consulting company or when another assessment system is used? Such research could shed more light on the experienced autonomy, competence and relatedness of employees in combination with the four components of the balanced scorecard.
5.6 Practical implications
The employees of the consultancy firm are highly educated and have a salary above average. This study assumes that the financial consultants are therefore intrinsically motivated.
Intrinsically motivated people would like to experience autonomy, competence and relatedness in work.
The impact of experienced relatedness on the valuation of (most) performance indicators is proven in this study. The most likely explanation for this relationship is that financial consultants who experience relatedness also feel related with the financial
consultancy company. When they feel related to the company, they also value the organization goals. The goals of the company are translated into the performance indicators.
It is very important for the financial consultancy company (or any company) that employees support the organization’s goals (and therefore the performance indicators). This study shows that the financial consultancy company can realize this by increasing the
experienced relatedness of the financial consultants with the company. The room for improving relatedness is probably quite high since relatedness of employees with the financial
37
relatedness can be done in many ways. For example, by organizing drinks, sharing knowledge, shared activities etcetera.
38
6 REFERENCES
Bandura, A. (1997). Self-efficacy: The exercise of control. New York: W. H. Freeman. Bernardin, H.J., Hagan, C.M., Kane, J.S., Villanova P. (1998) ‘Effective performance
management: a focus on precision, customers, and situational constraints’, Performance Appraisal: state of art in practice, San Francisco: Jossey-Bass Inc. pp. 3-48.
Boggiano, A. K., & Barrett, M. (1985). Performance and motivational deficits of helplessness: The role of motivational orientations. Journal of Personality and Social Psychology, 49, pp. 1753–1761.
Bonner, S.E., Sprinkle, G.B. (2002). The effects of monetary incentives on effort and task performance: theories, evidence, and a framework for research. Account., Organ. and Soc. 27, pp. 303–345.
Boxall, P. & Purcell, J. (2008). Strategy and human resource management. (second edition). Basingstoke: Palgrave Macmillan.
Bowlby, J. (1969). Attachment and loss: Vol. 1. Attachment. New York: Basic Books. Brown, M., & Benson, J. (2003). Rated to exhaustion? Reactions to performance appraisal
processes. Industrial Relations Journal, 34(1), pp. 67-81.
Cardy, R.L., Dobbins, G.H. (1994) Performance appraisal: alternative perspectives, Cincinatti: South-Western Publishing Co.
Carpentier, J., Mageau, G.A., (in press) When change-oriented feedback enhances motivation, well-being and performance: A look at autonomy-supportive feedback in sport, Psychology of Sport & Exercise (2013).
Cawley, D., Keeping, L.M. & Levy, P.E. (1998). Participation in the performance appraisal process and employee reactions: A meta-analytic review of field investigations. Journal of Applied Psychology, 83(4), pp. 615-633.
Challagalla, G. N. &Shervani, T. A., (1996). Dimensions and types of supervisorycontrol: effects on salesperson performance. Journal of Marketing, 60(1), pp. 89-106.
Cluitmans, J.J., (2002). M.m.v. Dekkers, M. A. F. en Van Oeffelt, T. Aan de slag met
Competenties: competentiegericht leren in HBO en MBO. Nuenen: Onderwijsadviesbureau Dekkers.
Daly, F., Teague, P., & Kitchen, P. (2003). Exploring the Role of Internal Communication during Organisational Change. Corporate Communications: An International Journal, 8(3), pp. 153-162.
