How Fit between the Performance Appraisal System
and Organizational Culture is related to Employee
Satisfaction with the Performance Appraisal System
Final thesis: June 26, 2014
Annemieke Koning (10296077)MSc. in Business Studies - Leadership & Management Track
Amsterdam Business School (UvA)
2
Table of Contents
Abstract
...4
1. Introduction
...5
2. Literature review
...7
2.1. Performance appraisal satisfaction ... 7
2.2. Organizational culture ... 9
2.3. Performance appraisal system and organizational culture... 12
2.4. Performance appraisal system – organizational culture fit ... 13
2.5. Direct and indirect fit ... 14
2.6. Hypotheses and conceptual model ... 15
3. Methods
...17
3.1. Research population ... 17
3.2. Research instruments and procedures ... 17
3.2.1. Organizational culture measure ... 18
3.2.2. Performance appraisal system measure ... 19
3.2.3. Performance appraisal system satisfaction measure ... 19
3.2.4. Direct performance appraisal system – organizational culture fit measure ... 20
3.3. Sample ... 20
4. Results
...21
4.1. Descriptive statistics and correlations ... 21
4.2. Hypotheses testing ... 23
4.2.1. Polynomial regression analysis and response surface analysis ... 24
4.3. Hypothesis 1 ... 24
4.3.1. Clan values ... 26
3 4.3.3. Market values ... 29 4.3.4. Hierarchy values ... 30 4.3.5. Summary hypothesis 1 ... 31 4.4. Hypothesis 2 ... 31 4.5. Hypothesis 3 ... 31 4.5.1. Clan values ... 33 4.5.2. Adhocracy values ... 34 4.5.3. Market values ... 35 4.5.4. Hierarchy values ... 36 4.5.5. Summary hypothesis 3 ... 37
5. Discussion
...37
5.1. Practical implications ... 405.2. Limitations and recommendations for future research ... 41
6. Conclusion
...43
7. Acknowledgements
...44
8. Reference list
...45
9. Appendices
...51
9.1. Appendix 1: Survey ... 51
9.2. Appendix 2: Complete correlation matrix ... 59
4
Abstract
Employee performance plays an important role in organizational success, which is why
organizations invest in performance management practices, such as performance appraisal
systems. However, employees are often unhappy with the performance appraisal system. This
study investigates to what extent fit between the performance appraisal system and
organizational culture is related to performance appraisal satisfaction as perceived by the
employee. Two types of fit were assessed in this regard: indirect (hypothesis 1) and direct fit
(hypothesis 2). Furthermore, this study looked into the relationship between indirect and
direct fit (hypothesis 3). The sample consisted of people from the employed working
population in The Netherlands (n = 178). For indirect fit, the four types of values as
determined by the Competing Values Framework were used. Hypotheses 1 and 3, involving
indirect fit, were tested using polynomial regression analysis and response surface analysis,
whereas the relationship between direct fit and performance appraisal satisfaction (hypothesis
2) was tested by means of a linear regression. The results showed support for all three
hypotheses. Clan and adhocracy values have the most explanatory power in regard to
performance appraisal satisfaction, having more impact when they are present in the
performance appraisal system rather than the organizational culture. The values of the
Competing Values Framework do not seem to capture the essence of what is it that employees
take into account when assessing direct fit. Furthermore, practical implications, and
limitations of the current study were discussed, and recommendations for future research were
5
1. Introduction
Steve Jobs once said "I make 50 cents for showing up (...) and the other 50 cents is based on
my performance” (Isaacson, 2011, p.479). However, when speaking of the exception to the rule, Steve Jobs may have been just that, as unfortunately, for most of the working population
this particular 50/50 situation does not seem to apply. The performance of an employee plays
a key role in organizational success (Pfeffer & Veiga, 1999); and fifty percent will not be
enough. Performance matters and it is for this reason that organizations invest in performance
management systems to manage the performance of their employees. Performance
management is a ‘continuous process of identifying, measuring and developing the
performance of organization members and aligning performance with the strategic goals of
the organization’ (Aguinis, 2007, in Biron, Farndale & Paauwe, 2011, p. 1295).
The pressure of the current economy is forcing organizations to be more efficient and
effective and therefore emphasis on performance management is more important than ever.
However, although organizations might see the importance of performance management, the
subjects of the performance management (i.e. the employees) often do not. Whereas almost
every organization implements some type of system to evaluate the performance of their
employees (Aguinis, Joo & Gottfredson, 2011; Cascio, 2006), employees are often unhappy
with these performance management practices, especially in regard to performance appraisal
(Abu-Doleh & Weir, 2007; Molleman & Timmerman, 2003; Noe, Hollenbeck, Gerhart &
Wright, 2010; Zedeck, 2011). This is worrisome, as it is suggested that employee opinions of
the performance appraisal system are equally as important as the system’s validity and
reliability (Lawler, 1967; Hedge & Teachout, 2000). Organizations should pay attention to
performance appraisal and taking an employee perspective may offer the organization an
6 Not only is the working environment changing due to current economic conditions and
the pressure that this brings along, the context of organizations is changing as well.
Traditional ways of working are increasingly making way for new ways of working,
supported by new technology. The latter makes it possible to work outside the office and
currently in The Netherlands almost 1 out of 3 workers is working from home during a part of
the time (Centraal Bureau voor de Statistiek [CBS] & TNO, 2012). Also, the number of
organizations making their office facilities remote is growing rapidly (CBS, 2013a). These
new ways of working are an important issue in today’s organizations and will only become more visible as generation Y begins to take over. It is clear that the organizational context is
changing.
Will this have implications for performance management practices adopted by the
company? From the definition of performance management described above it can be derived
that strategic and organizational factors permeate performance management (Aguinis, 2007,
in Biron et al., 2011). Performance management practices cannot be separated from the
organizational context. However, although organizations often implement the most
sophisticated performance management practices, they often fail to take into account the
organizational context (Haines & St. Onge, 2012, p.1169). What will this mean for the
effectiveness regarding positive organizational outcomes of performance management
practices? How important is it that performance management practices and the organizational
context are aligned? The aim of the current study is to try and find an answer to these
questions, with a specific focus on performance appraisal, which is a very important aspect of
performance management (Noe et al., 2010).
This paper will be structured as follows. First, having already introduced the topic
shortly in the above section, a more detailed literature review will describe and discuss the
7 research methods will be described, followed by an overview of the results. The paper will be
concluded by a discussion and conclusion, in which the results, practical implications, and
limitations of the study, as well as suggestions for future research will be discussed.
2. Literature Review
2.1. Performance appraisal satisfactionPerformance appraisal is a part of the broader process of performance management and is
described as ‘the process through which an organization gets information on how well an employee is doing his or her job’ (Noe et al., 2010, p. 351; Zedeck, 2011). Performance
appraisal is described as one of the most important human resource practices and has to be
performed correctly in order to achieve competitive advantage (Boswell & Boudreau, 2002;
Noe et al., 2010). According to Noe et al. (2010), performance management can serve several
purposes: strategic, administrative, and developmental. The performance appraisal system
plays an important role in all three purposes and perhaps that is why performance appraisal is
one of the most researched topics within the field of performance management (Fletcher,
2001). When considering the outcomes of performance appraisal, often outcomes are
measured on an organizational level (e.g. influence of performance appraisal on
organizational performance). However, the reactions of the ratees (i.e. the employees) should
not be disregarded (Keeping & Levy, 2000; Zedeck, 2011). Perceptions of performance
appraisal may influence people’s judgment and attitudes towards the process, and in turn
influence organizational performance (Purcell & Hutchinson, 2007). This particular aspect of
performance appraisal research focuses on employee reaction to performance appraisal.
Dissatisfaction with the performance appraisal systems in place may signal a lack of success
8 suggested that favorable employee perceptions of performance appraisal are needed in order
to positively influence employee behavior and future development (Fletcher, 2001). The most
frequently measured appraisal reaction is performance appraisal satisfaction (Giles &
Mossholder, 1990; Keeping & Levy, 2000), which suggests that it is important. Levy and
Williams (2004) pointed out that although a lot of research is done on factors that may
contribute to performance appraisal satisfaction, a lack of empirical evidence exists on the
why and how performance appraisal satisfaction matters and more field research on this
subject is needed. The empirical research that has been done, however, suggests a positive
association between performance appraisal satisfaction and various positive work outcomes.
In a longitudinal study, Blau (1999) showed performance appraisal satisfaction to influence
overall job satisfaction. Years later, an empirical study by Kuvaas (2006) showed that
performance appraisal satisfaction was directly related to affective commitment as well as
intention to leave, and indirectly to work performance, providing empirical support that
performance appraisal satisfaction may lead to positive work outcomes.
Performance appraisal satisfaction can be divided into two components: satisfaction
with the appraisal system and satisfaction with the review session, hereafter referred to as
system satisfaction and session satisfaction respectively (Giles & Mossholder, 1990). System
satisfaction concerns satisfaction with those aspects of the performance appraisal system that
primarily exist independently of the actions of the raters, making it objective (Giles, Findley
& Feild, 1997). Session satisfaction on the other hand, can be found on the more subjective
end of the spectrum (Giles et al., 1997). The current study will only focus on the objective
system satisfaction component, as it was not possible within the boundaries of this study to
9 2.2. Organizational culture
The environment in which people work is important. A work environment variable that can
have a major impact on the members of the organization is organizational culture.
Organizational culture is a ‘pattern of basic assumptions, invented, discovered, or developed by a given group, as it learns to cope with its problems of external adaptation and internal
integration, that has worked well enough to be considered valid and, therefore is to be taught
to new members as the correct way to perceive, think, and feel in relation to those problems’
(Schein, 1990, p.111). Although there is a widespread claim among many researchers that
organizational culture is linked to the performance of the organization, there seems to be no
consensus regarding the mechanisms through which this linkage occurs. For instance, some
researchers claim that the strength (i.e. how widely the values of the organization are shared
within the organization) of the culture is key (e.g. Smart & John, 1996), whereas others adapt
the resource based view, stating that the organizational culture should be valuable, rare,
inimitable, and non-substitutable in order for an organization to be successful (Barney, 1991;
Ogbonna & Harris, 2000). Regardless of the lack of universality in regard to the mechanism
that links organizational culture and performance, the evidence does generally highlight the
importance of organizational culture within an organization (Ogbonna & Harris, 2000).
In regard to the measurement of organizational culture, an extensive amount of
different scales and models exist (Scott, Mannion, Davies & Marshall, 2003; Yu & Wu,
2009). This study will adopt the Competing Values Framework, developed by Cameron and
Quinn (2011), based on several reasons. First of all, in order to determine which measurement
scale is appropriate, it is important that the measurement captures the essence of the various
organizational culture aspects (Ralston, Terpstra-Tong, Terpstra, Wang & Egri, 2006). In an
extensive review of literature, Detert, Schroeder and Mauriel (2000) identified eight common
10 dimensions to a certain degree (Ralston et al., 2006). Secondly, the Competing Values
Framework is the measurement that is most often used in quantitative research concerning
organizational culture (Kwan & Walker, 2004). Thirdly, the focus of the framework is on the
individuals that comprise the organization (Cameron & Quinn, 2011), which is in line with
the focus of this study. Another important factor in the choice for this particular model is that
a large amount of empirical research has established the validity and reliability of this
framework (Yu & Wu, 2009).
According to the Competing Values Framework, organizational culture varies along
two scales: in terms of focus (i.e. internal or external) and degree of control (i.e. flexibility or
stability), which results in four types of organizational culture: clan, adhocracy, market, and
hierarchy (Quinn, 1988, in Hooijberg & Petrock, 1993; Cameron & Quinn, 2011). Clan
culture can be described as a very personal place, it is like an extended family. Collaboration
and participation are key in this culture, and the leaders fulfill a mentor role. Adhocracy
culture is characterized by a dynamic and entrepreneurial workplace in which people are
willing to take risks. Innovation and experimentation are important. Market culture is
results-oriented, it is about getting the job done. Competition and achievement are very important in
this culture. Hierarchy culture is about structure and control. In this, coordination and
efficiency are important, as well as following the rules and procedures (Cameron & Quinn,
11 Figure 1: Competing Values Framework (retrieved from: Cameron & Quinn, 2011, p. 39).
The type of organizational culture that is dominant within the organization is determined by
using six key dimensions: dominant characteristics, leadership, the management style, means
of bonding, strategic plans, and criteria of success. The first dimension, dominant
characteristics of the organization, describes what the overall organization is like. The second
dimension is about the organizational leadership style and approach that are present in the
organization. Thirdly, the management style dimension refers to the style that characterizes
how employees are treated and what the working environment is like. The fourth dimension
describes the means used by the organization to hold the organization together. The fifth
dimension is concerned with the drivers of organizational strategy. The criteria of success, the
sixth and final dimension, determine what is important to the organization in terms of what
12 2.3. Performance appraisal system and organizational culture
Although organizational contextual factors have always been recognized as being important
(Jackson & Schuler, 1995) not much empirical research into the effect of contextual factors on
performance management has been done (Haines & St. Onge, 2012; Levy & Williams, 2004).
Cawley, Keeping and Levy (1998) suggest that the performance appraisal system is dependent
on and must be consistent with the larger organizational context. Moreover, it is suggested
that because of interdependency, ignoring the impact of organizational culture on
organizational practices such as performance appraisal could even result in counterproductive
outcomes (Magee, 2002, in ul Mujeeb & Ahmad, 2011). This highlights the importance of
examining the relationship between organizational culture and performance appraisal into
more detail (Fletcher, 2001). Taking into account the statement made by Posthuma and
Campion (2008, in Haines & St. Onge, 2012, p. 1159) that “too much attention has been
placed on the design of a [performance management] system, and not enough on how it works
when implemented”, Haines and St. Onge (2012) proposed that a collaborative organizational culture (i.e. clan culture) will lead to more positive performance management outcomes, as
this type of organizational culture will value employee engagement. However, the authors did
not find evidence supporting their statement that this particular type of organizational culture
moderates the relationship between performance management practices and effectiveness.
This does not mean that the impact of organizational culture on performance management
outcomes should be disregarded, but a different perspective may offer better insight. Overall,
there seems to be a consensus that performance management practices should be congruent
with the culture of the organization (Aguinis et al., 2011). For example, in an organization
with a dominant hierarchic organizational culture, it is very likely that a 360-degree feedback
13 work with, will not result in positive organizational outcomes, as this will not ‘fit’ together
(Aguinis et al., 2011).
This suggestion is in line with contingency theory, that describes organizational
performance as resulting from a fit between characteristics of the organization and
environmental aspects (Fredericks, 2005). More specifically, in regard to human resource
management, it is about ensuring that human resource practices are appropriate for the
circumstances of the organization, including the culture (Delery & Doty, 1996; Wright,
Gardner, Moynihan & Allen, 2005). Drawing on contingency theory, outcomes of the
performance appraisal system will depend on the fit between the performance management
system and the organizational culture (Chenhall, 2003; Gerdin & Greve, 2008). Whether this
will be the case in relation to the outcome of employee satisfaction with the performance
appraisal system is the subject of this study.
To what extent is fit between the performance appraisal system and organizational
culture related to employee satisfaction with the performance appraisal system?
2.4. Performance appraisal system – organizational culture fit
In order to assess the fit between the performance appraisal system and the organizational
culture, it is important to determine what fit entails exactly. Although there seems to be a
consensus in regard to the importance of performance appraisal system – organizational
culture fit, no previous empirical research on this concept is available to build on. Therefore,
this study will borrow from a research field that is somewhat related: person-organization fit.
In this particular field, extensive research has been done and person-organization fit is
determined in various ways. One of these methods consists of assessing value congruence
between the person and the organization. Value congruence is considered the most consistent and effective predictor of employee outcomes (Westerman & Cyr, 2004). Measuring values of
14 employee and organization along the same scale is a so called commensurate measure
(Schneider, Goldstein & Smith, 1995). This type of measure allows to assess fit between the
performance appraisal system and the organizational culture by directly comparing the two
variables, and thus achieve high predictive validity. Furthermore, value congruence can be
combined with the values as described by the Competing Values Framework (Kalliath,
Bluedorn & Strube, 1999; van Vuuren, Veldkamp, de Jong & Seydel, 2007). Hence, although
it is recognized that person-organization fit and performance appraisal system-organizational
culture fit differ from one another in some aspects, this study will adopt value congruence
from the field of person-organization fit as a method to assess the fit in the current study.
2.5. Direct and indirect fit
Also drawing on the field of person-organization fit, a distinction can be made between
assessing fit directly and indirectly. Direct fit basically entails asking a person to judge the
extent to which they believe there is a fit between two variables, in this case between the
performance appraisal system and the organizational culture. Because of the subjective nature,
this type of fit is also referred to as perceived fit (Kristof, 1996). Indirect fit, on the other
hand, is also referred to as actual fit and can be described as the empirical relationship
between separate assessments of the underlying values of the two variables (Kristof, 1996).
Within this type of fit, a distinction can be made between subjective and objective indirect
assessment of fit. Indirect subjective fit compares the values of the performance appraisal
system with the values of the organizational culture as perceived by the employee, whereas
indirect objective fit compares the values of the performance appraisal system with the values
of the organizational culture as determined by other sources (Kristof-Brown, Zimmerman &
Johnson, 2005). In the person-organization fit field there seems to be little consensus on
15 measures of fit in order to better assess the relative effect of each (Kristof, 1996). This study
will follow this recommendation, insofar that it will assess both the direct fit, and the indirect
subjective fit. Due to the time and resource constraints linked to this particular study, it was
decided not to include indirect objective fit. It has to be noted that from this point forward,
indirect subjective fit will be simply referred to as indirect fit, because indirect objective fit is
not included in this study.
2.6. Hypotheses and conceptual model
The purpose of this study is to analyze the influence of the fit between the performance appraisal system and the organizational culture, which is operationalized as value congruence between the performance appraisal system and the organizational culture, on employee satisfaction with the performance appraisal system, using direct as well as indirect fit measures.
Based on contingency theory (Fredericks, 2005), as well as general consensus that
management practices should be congruent with the culture of the organization in order to
achieve favorable organizational outcomes (Aguinis et al., 2011), it is expected that this will
also be the case in relation to the performance appraisal system, for both direct and indirect
fit.
Hypothesis 1: A better indirect fit between the performance appraisal system and the organizational culture will be positively related to satisfaction with the performance appraisal
system.
Hypothesis 2: A better direct fit between the performance appraisal system and the organizational culture will be positively related to satisfaction with the performance appraisal
system.
Furthermore, although both direct and indirect fit are hypothesized to lead to more
satisfaction towards the performance appraisal system, it is expected that a relationship
16 based on perceptions of the employee, and therefore not objective in nature, it can be assumed
that indirect fit is relatively more objective than direct fit. This is based on the fact that
indirect fit captures the perception of values that were operationalized independently from the
employee (i.e. by the researcher and the Competing Values Framework) and can therefore be
called objective. Assuming that the more objective perceptions of these operationalizations
will to a certain extent influence the more subjective perceptions, it is expected that the
indirect fit between the performance appraisal system and the organizational culture will
influence how this fit is perceived directly by the employee.
Hypothesis 3: A better indirect fit between the performance appraisal system and the organizational culture will be positively related to the direct fit between the performance
appraisal system and the organizational culture.
Figure 2 shows the conceptual model, and the hypotheses.
17
3. Methods
3.1. Research population
This study was conducted among the employed working population of The Netherlands,
which, translated from the Dutch definition, consists of ‘all people in the age from 15 till 65 who live in The Netherlands and have paid employment for twelve or more hours per week’ (CBS, 2013b). This population consists of approximately 7.2 million people, of which
approximately 1.1 million are self-employed (CBS - StatLine, 2013a). As people who are
self-employed are not the subject of a performance appraisal, they were not eligible for the
current study and therefore excluded, leaving a research population of approximately 6.1
million people. No sampling frame could be retrieved for this research population, therefore a
non-probability sampling method was used. The main goal was to achieve a sample as large
as possible, which would increase the chance of having a representative sample and being
able to generalize research outcomes to the entire research population.
3.2. Research instruments and procedures
The data were collected through an online survey, which was completely anonymous. First of
all, the questionnaire was distributed within the researcher’s personal network (i.e. primary
respondents) by means of social media and email. Then, snowball technique was used to
reach further respondents, asking the primary respondents to spread the questionnaire in their
personal network. Using this method provided the opportunity to reach a large group of
people. Unfortunately, it also made it impossible to provide an exact response rate, as it is
unknown how many respondents were reached beyond the primary network. There was no
18 The questionnaire started with four questions to determine whether someone was
eligible or not to participate in the study (i.e. whether they are a part of the research
population). For example, they were asked if they live in The Netherlands. If eligible,
respondents were forwarded to the research part of the questionnaire, and if not, they were
forwarded to a thank you message. The research part of the questionnaire contained questions
regarding the independent (i.e. organizational culture, performance appraisal system, and
direct fit) and dependent (i.e. performance appraisal satisfaction) variables, as well as a
number of demographical questions, such as age (see Appendix 1 for the complete
questionnaire). Indirect fit between the performance appraisal system and organizational
culture was calculated using polynomial regression after the data were collected (Edwards &
Parry, 1993; Shanock, Baran, Gentry, Pattison & Heggestad, 2010).
3.2.1. Organizational culture measure
Organizational culture is a ‘pattern of basic assumptions, invented, discovered, or developed
by a given group, as it learns to cope with its problems of external adaptation and internal
integration, that has worked well enough to be considered valid and, therefore is to be taught
to new members as the correct way to perceive, think, and feel in relation to those problems’
(Schein, 1990, p.111). Organizational culture was assessed with the Organizational Culture
Assessment Instrument (OCAI) that consists of 24 items total (Cameron & Quinn, 2011). The
OCAI differentiates four culture types, which were measured separately: clan, adhocracy,
market, and hierarchy. Reliability analyses showed a Cronbach’s alpha of .86, .91, .89, and
.77, respectively. An example item of the OCAI scale describing clan culture is “The
management style in the organization is characterized by teamwork, consensus, and
19 3.2.2. Performance appraisal system measure
The performance appraisal system was assessed by means of a self-developed scale based on
the values and dimensions of the OCAI items. The performance appraisal system scale
consists of 24 items and was developed with the help of subject matter experts, as well as
respondents representative of the intended sample. In line with the organizational culture
measure, the performance appraisal system scale consisted of four subscales measuring the
clan, adhocracy, market, and hierarchy values of the performance appraisal system. Reliability
analyses showed a Cronbach’s alpha of .86, .85, .77, and .67 respectively, with the latter value
being debatable, as the rule of thumb indicates that alpha should be higher than .70. In regard
to this issue, the reliability analysis showed that the first item did not correlate well with the
composite score from the other five items of the hierarchy subscale (i.e. item-total correlation
was .19, which is well below the proposed number of .30), and that removing the first item
would increase reliability to an alpha of .71, providing clear statistical reason to delete the
first hierarchy item from this scale. This statistical reason was in this case deemed sufficient
to decide to remove the item, because of the fact that the scale was self-developed, and has
not been validated before. Analysis was therefore continued with a 23 item-scale.
Furthermore, an exploratory factor analysis was conducted, which, in combination with a
thorough investigation of the content of the items, did not provide sufficient reason to remove
any more items. An example item, translated from the Dutch item, of the performance
appraisal system scale is “I feel that the performance appraisal system is focused on
teamwork, consensus, and participation”.
3.2.3. Performance appraisal system satisfaction measure
Performance appraisal system satisfaction was assessed by means of the three-item System
20 satisfaction with those aspects of the performance appraisal system that primarily exist
independently of the actions of the raters (Giles et al., 1997). Reliability analysis showed a
Cronbach’s alpha of .90. An example item of the System Satisfaction scale (1990) is “In general, I feel the company has an excellent performance appraisal system”.
3.2.4. Direct performance appraisal system – organizational culture fit measure
Direct fit between the performance appraisal system and organizational culture was assessed
by means of an adapted version of the Person-Organization fit scale (Cable & DeRue, 2002).
The adapted scale describes an individual’s overall judgment about the extent to which he or
she perceives the performance appraisal system and organizational culture to be congruent in
terms of underlying values (Kristof, 1996). Reliability analysis showed a Cronbach’s alpha of
.85. An example item of the adapted Cable & DeRue scale for direct performance appraisal
system-organizational culture fit (2002) is “The values of the performance appraisal system
match my organization’s values and culture.”.
For all measures, items were rated on a Likert scale ranging from 1 (strongly disagree) to 7
(strongly agree) and translated from English to Dutch (with exception of the self-developed
Performance Appraisal System scale, which was developed in Dutch). For the translation a
method of forward- and back-translation was used to avoid a loss in content validity. The
complete questionnaire in Dutch can be retrieved from Appendix 1.
3.3. Sample
The final sample consisted of 178 respondents, of which 75 were men (42.1%) and 103 were
women (57.9%). The mean age of the respondents was 40.50 years (SD = 12.22, MIN = 21,
MAX = 65). The average tenure with the current employer was 7.35 years (SD = 6.69; MIN =
21 followed by higher professional education (37.1%), senior secondary vocational education
(7.1 %), and secondary education (1.7%). Respondents were employed in a wide range of
sectors, with the majority of the respondents working in business services (40.4%). Other
sectors that provided a substantial amount of the respondents were ‘other services’ (21.3%), government organizations (14.0%), and health care (9.6%).
Comparing the abovementioned sample descriptives to CBS statistics of the research
population provides a preliminary insight into the representativeness of the sample. In terms
of average age, the mean of the sample (40.50 years) lies very close to the mean of the
research population, which is 41.70 years (CBS - StatLine, 2013b). In regard to gender, the
sample is slightly reversed in the male/female distribution, as the research population consists
of slightly more men (52.5%) than women (47.5%), whereas the sample consists of more
women than men (CBS - StatLine, 2013a). Education wise, approximately 36% of the
research population has either finished university or higher professional education, and
approximately 43% has completed senior secondary vocational education or secondary
education (CBS - StatLine, 2013a). It is clear that in terms of education level, the research
population statistics do not match the sample.
4. Results
4.1. Descriptive statistics and correlations
Table 1 contains the mean scores, standard deviations, correlations, and reliability coefficients
of the independent and dependent variables that were measured in the survey. The complete
22 Table 1: Means, standard deviations, correlations, and reliability coefficients
Variables M SD 1 2 3 4 5 6 7 8 9 10 1. PA system - Clan 4.48 1.19 (.86) 2. PA system - Adhocracy 3.90 1.17 .71** (.85) 3. PA system - Market 4.53 1.00 .22** .33** (.77) 4. PA system -Hierarchy 4.33 1.02 -.01 .00 .33** (.71) 5. OC - Clan 4.37 1.19 .65** .48** .05 -.06 (.86) 6. OC - Adhocracy 4.23 1.33 .51** .56** .22** -.19* .64** (.91) 7. OC - Market 4.33 1.33 .14 .15* .67** .12 .11 .38** (.89) 8. OC - Hierarchy 4.31 1.05 .01 -.13 .19* .52** .12 -.10 .31** (.77) 9. Direct PA system – OC fit 4.93 1.11 .32** .23** .22** .12 .35** .27** .27** .26** (.85) 10. PA satisfaction 4.49 1.37 .67** .54** .24** .07 .56** .42** .28** .20** .43** (.91)
PA = performance appraisal; OC = organizational culture ** Correlation is significant at the 0.01 level (2-tailed). * Correlation is significant at the 0.05 level (2-tailed).
Looking at table 1, it is clearly visible that almost all correlations show a significant positive
result. In regard to the measurement of the four types of values of the performance appraisal,
(i.e. variable 1 to 4), as well as the four types of organizational culture (i.e. variable 5 to 8),
this is not entirely surprising, because the four quadrants of the Competing Values Framework
are connected to each other (see figure 1, p. 11). For instance, in terms of degree of control,
clan and adhocracy values are both flexible, and in terms of focus, clan and hierarchy values
share an internal orientation. The significant correlations in table 1 that may be explained by
the overlap in terms of degree of control or focus between the four types of values of the
23 .33), and market and hierarchy (r = .33). In regard to the organizational culture types, this
overlap may explain the significant correlation between clan and adhocracy (r = .64), and
market and hierarchy (r = .31). Taking this into account, this leaves one significant correlation
that cannot be explained. Clan and market values show a positive correlation in regard to the
four types of values of the performance appraisal system (r = .22).
Furthermore, is stands out in table 1 that each of the four types of organizational
culture correlates positively and significantly with the value of the performance appraisal
similar to it (clan r = .65; adhocracy r = .56; market r = .67; hierarchy r = .52). Other
significant correlations regarding the organizational culture types and values of the
performance appraisal system exist between clan and adhocracy (r = .48), adhocracy and clan,
market, hierarchy (r = .51; r = .22; r = -.19), market and adhocracy (r = .15), hierarchy and
market (r =.19), respectively.
With the exception of the hierarchy values of the performance appraisal system, there
were significant positive correlations between the values of the performance appraisal system
as well as the types of organizational culture, and the direct fit between the performance
appraisal system and organizational culture (see table 1). Similar to this, all values of the
performance appraisal system, with the exception of the hierarchy values, and all
organizational culture types correlated positively and significantly with performance appraisal
satisfaction (see table 1). Direct fit between the performance appraisal system and
organizational also correlated positively and significantly with performance appraisal
satisfaction (r = .43).
4.2. Hypotheses testing
Hypotheses 1 and 3 were tested by means of polynomial regression and response surface
24 hypotheses testing will be discussed, polynomial regression and response surface analysis will
be explained further, as this method is not as well-known as linear regression.
4.2.1. Polynomial regression analysis and response surface analysis
Polynomial regression analysis provides an opportunity to examine the extent to which a
combination of two independent variables relate to the dependent variable (Shanock et al.,
2010). This type of analysis has been used before in various fit studies somewhat similar to
the current study, also including the four types of values as described by the Competing
Values Framework (Cameron & Quinn, 2011; Kalliath et al., 1999; van Vuuren et al., 2007).
Polynomial regression was proposed by Edwards (1993, 1994) as an alternative to the
traditional congruence tests, such as difference scores, and offers a solution for the numerous
methodological flaws that come with those particular tests. For example, difference scores
combine two measures into one score, and by doing this, do not offer insight into the extent to
which each of the components contribute to the outcome, whereas polynomial regression does
provide this option (Shanock et al., 2010).
Response surface analysis was conducted to aid interpretation of the results. For this,
the Excel spreadsheet made available by Shanock et al. (2010) was used, calculating the
slopes and curvatures and providing three-dimensional plots of the variables involved in the
hypotheses. Furthermore, the article by Shanock et al. (2010) was used as a guide to interpret
the results.
4.3. Hypothesis 1
Hypothesis 1 was tested with the use of polynomial regression analysis and response surface
analysis. Table 2 shows the explained variance (R2) of the polynomial regression analyses.
25 Performance appraisal satisfaction = X + Y + X2 + XY + Y2
In this equation X represented the underlying values of the performance appraisal system (e.g.
clan values of the performance appraisal system), and Y represented the type of
organizational culture (e.g. clan culture).
Table 2: R2 and significance (p) of the regressions of performance appraisal system values and organizational culture values in relation to performance appraisal satisfaction.
R2 Sig. (p)
Clan .531 .000
Adhocracy .355 .000
Market .169 .000
Hierarchy .069 .030
In line with the hypothesis, it was expected that the highest level of performance appraisal
satisfaction would be reached when there is perfect agreement between the values of the
performance appraisal system and the organizational culture, so when X equals Y. This line of
perfect fit X=Y should be straight and horizontal, because the level of value agreement should
not influence the level of the outcome value (e.g. a Likert scale score of five for both X and Y
is expected to show the same level of outcome value as a Likert scale score of three for X and
Y). The line of misfit (X = -Y) should show an inverted U-shape, as it is expected that the
outcome value will decrease when there is a higher degree of discrepancy between X and Y.
For every regression equation of the four in total, slopes and curvatures of the regression
26 Table 3: Performance appraisal satisfaction as predicted by indirect PA system – OC fit.
Clan Adhocracy Market Hierarchy
Slope X = Y 0.86** 0.60** 0.59** 0.25* Curvature X = Y -0.11* -0.12* -0.31** -0.15 Slope X = -Y 0.48** 0.32* 0.26 -0.33 Curvature X = -Y -0.26* -0.20 -0.64** -0.31 * p < .05, ** p < .01 4.3.1. Clan values
The regression of the clan values and performance appraisal satisfaction shows a significant
result, as well as reasonably high explained variance (R2 = 0.531, see table 2). The graph of
this regression is shown in figure 3.1.
Figure 3.1: PA satisfaction as predicted by the indirect fit of the clan values of the performance appraisal system and the organizational culture. Note that PA system and OC values were centered.
Table 3 shows that the slope as well as the curvature of the line of fit (X = Y) were
significant. The slope of X = Y is positive, indicating that performance appraisal satisfaction
P
A Satis
fa
cti
27 increases as the clan values of the performance appraisal system and organizational culture
increase. The significance of the curvature of the fit line shows that this relationship is
non-linear. As this curvature is negative, this means that there is a downward curving of the fit
line. This is visible in figure 3.1.
In regard to the misfit (X = -Y) line, also both the slope and curvature showed
significant results, indicating the proposed inverted U-shape, which is somewhat visible in
figure 3.1. However, as can be seen in the graph, this U-shape is not perfectly balanced. The
negative significant curvature of the misfit line (X = -Y) indicates that performance appraisal
satisfaction decreases more sharply as the discrepancy between the clan values of the
performance appraisal system and the organizational culture increases. The value of the slope
of X = -Y dictates the direction of the discrepancy. The positive significant value of this slope
indicates that performance appraisal satisfaction is affected more when the clan values of the
performance appraisal system are higher than the organizational culture, than when the
opposite is the case.
4.3.2. Adhocracy values
The regression of the adhocracy values and performance appraisal satisfaction shows a
significant result, as well as reasonably high explained variance (R2 = 0.355, see table 2). The
28 Figure 3.2: PA satisfaction as predicted by the indirect fit of the adhocracy values of the performance appraisal system and the organizational culture. Note that PA system and OC values were centered.
Table 3 shows that the slope as well as the curvature of the fit line (X = Y) are significant.
The slope of the fit line has a positive coefficient, indicating that in this case performance
appraisal satisfaction increases as the adhocracy values of the performance appraisal system
and organizational culture increase. The significance of the curvature of the fit line indicates
that this relationship is non-linear, and because the curvature has a negative coefficient, this
means that there is a downward curving of the fit line. This is supported by the graph in figure
3.2.
In regard to the line of misfit (X = -Y), only the slope showed a significant result. The
positive significant value of this slope indicates that performance appraisal satisfaction is
affected more when the adhocracy values of the performance appraisal system are higher than
the organizational culture, than when the opposite is the case.
P
A Satis
fa
cti
29 4.3.3. Market values
The regression of the market values and performance appraisal satisfaction shows a
significant result, but the explained variance was relatively weak (R2 = 0.169, see table 2).
The graph of this regression is shown in figure 3.3.
Figure 3.3: PA satisfaction as predicted by the indirect fit of the market values of the performance appraisal system and the organizational culture. Note that PA system and OC values were centered.
Both the slope and the curvature of the fit line (X = Y) show significant results (see table 3).
The slope has a positive coefficient, which indicates that in this case performance appraisal
satisfaction increases as the market values of the performance appraisal system and
organizational culture increase. The significance of the curvature of the fit line indicates that
this relationship is non-linear, and because the curvature has a negative coefficient, this means
that the line of fit shows a downward curving, as is visible in figure 3.3.
In regard to the line of misfit (X = -Y), only the curvature showed a significant result
(see table 3). This result has a negative coefficient, indicating that when the discrepancy
between the market values of the performance appraisal system and the organizational culture
P
A Satis
fa
cti
30 increases, the performance appraisal satisfaction will decrease more sharply. This is also
clearly visible in figure 3.3.
4.3.4. Hierarchy values
The regression of the hierarchy values and performance appraisal satisfaction shows a
significant result, but the explained variance was very weak (R2 = 0.069, see table 2). The
graph of this regression is shown in figure 3.4.
Figure 3.4: PA satisfaction as predicted by the indirect fit of the hierarchy values of the performance appraisal system and the organizational culture. Note that PA system and OC values were centered.
Only the slope of the line of fit (X = Y) showed a significant result, indicating a straight line
of fit. As visible in figure 3.4, this line rises as the value of fit is higher. The positive
coefficient of the slope supports this notion.
P
A Satis
fa
cti
31 4.3.5. Summary hypothesis 1
The regression analyses (see table 2) supported hypothesis 1, for each of the four types of
values. However, for all four value types, the fit lines did not show the straight and horizontal
shape that was expected. In all four cases, the fit line rose as the level of the values got higher.
Also, for the clan, adhocracy, and market values, the fit line was not linear. Furthermore,
coefficients indicate that in regard to clan and adhocracy values, the performance appraisal
system has a stronger effect on performance appraisal satisfaction than organizational culture.
4.4. Hypothesis 2
Hypothesis 2 was tested with the use of linear regression. Table 1 shows that there is a
significant positive correlation (r = .43, p < 0.01) between direct performance appraisal
system – organizational culture fit and performance appraisal satisfaction. This direct fit
explained approximately 18% (R2 = 0.182) of the variance in performance appraisal
satisfaction and was found to be a significant predictor of performance appraisal satisfaction.
This result supports hypothesis 2: a better direct fit between the performance appraisal system
and the organizational culture is positively related to satisfaction with the performance
appraisal system. In other words, when employees perceive a better fit between the values of
the performance appraisal system and the organizational culture, they will be more satisfied
with the performance appraisal system.
4.5. Hypothesis 3
Hypothesis 3 was tested with the use of polynomial regression analysis and response surface
analysis. Table 4 shows the explained variance (R2) of the polynomial regression analyses.
The general regression equation was as follows:
32 In this equation X represented the underlying values of the performance appraisal system (e.g.
clan values of the performance appraisal system), and Y represented the type of
organizational culture (e.g. clan culture).
Table 4: R2 and significance (p) of the regressions of performance appraisal system values and organizational culture values in relation to the direct fit.
R2 Sig. (p)
Clan .155 .000
Adhocracy .112 .001
Market .123 .000
Hierarchy .109 .001
Similar to hypothesis 1, hypothesis 3 also predicted that the highest level of the outcome
variable (here: perceived direct fit) would be reached when there is perfect agreement
between the values of the performance appraisal system and the organizational culture, so
when X equals Y. This line of perfect fit X = Y should be straight and horizontal, because the
level of value agreement should not influence the level of the outcome value (e.g. a Likert
scale score of five for both X and Y is expected to show the same level of outcome value as a
Likert scale score of three for X and Y). The line of misfit (X = -Y) should show an inverted
U-shape, as it is expected that the outcome value will decrease when there is a higher degree
of discrepancy between X and Y. For every regression equation of the four in total, slopes and
curvatures of the regression equations were calculated. These calculated values are shown in
33 Table 5: Direct PA system – OC fit as predicted by indirect PA system – OC fit.
Clan Adhocracy Market Hierarchy
Slope X = Y 0.37** 0.32** 0.41** 0.21* Curvature X = Y 0.09 0.11* -0.09 0.20** Slope X = -Y -0.02 -0.04 0.39 -0.27 Curvature X = -Y -0.01 0.06 -0.53** 0.00 * p < .05, ** p < .01 4.5.1. Clan values
The regression of the clan values and direct fit shows a significant result, but the explained
variance was relatively weak (R2 = 0.155, see table 4). The graph of this regression is shown
in figure 4.1.
Figure 4.1: Direct PA system – OC fit as predicted by the indirect fit of the clan values of the performance appraisal system and the organizational culture. Note that PA system and OC values were centered.
Dire ct P A s y stem – OC f it
34 In regard to the clan values, only the slope of the fit line (X = Y) showed a significant result
(see table 5). Although figure 4.1 shows that the fit line is quite horizontal, there is a slight
rise visible as well. The positive coefficient of the slope supports this notion, indicating that
when the clan values of the performance appraisal system and the organizational culture were
in agreement, direct fit increased as the level of the clan values of the performance appraisal
system and the organizational culture increased. The non-significant result of the curvature of
the fit line indicates that this relationship is linear.
4.5.2. Adhocracy values
The regression of the adhocracy values and direct fit shows a significant result, but the
explained variance was relatively weak (R2 = 0.112, see table 4). The graph of this regression
is shown in figure 4.2.
Figure 4.2: Direct PA system – OC fit as predicted by the indirect fit of the adhocracy values of the performance appraisal system and the organizational culture. Note that PA system and OC values were centered.
Somewhat similar to figure 4.1, figure 4.2 shows the fit line (X = Y) to be quite horizontal, as
was predicted. However, as the slope of the fit line is significant and has a positive
D ire ct P A s y ste m – O C f it
35 coefficient, this indicates that direct fit between the performance appraisal system and the
organizational culture increases as the levels of adhocracy values of the performance appraisal
system and the organizational culture increase. The significant curvature of the fit line
indicates that the relationship between the independent and dependent variables is non-linear.
In line with the positive coefficient of the curvature, figure 4.2 shows a slight upward curving
of the fit line.
4.5.3. Market values
The regression of the market values and direct fit shows a significant result, but again the
explained variance was relatively weak (R2 = 0.123, see table 4). The graph of this regression
is shown in figure 4.3.
Figure 4.3: Direct PA system – OC fit as predicted by the indirect fit of the market values of the performance appraisal system and the organizational culture. Note that PA system and OC values were centered.
The slope of the fit line (X = Y) shows a positive significant result. This means that the direct
fit between the performance appraisal system and the organizational culture increases as the
levels of market values of the performance appraisal system and the organizational culture
Dire ct P A s y stem – OC f it
36 increase. The non-significant result of the curvature of the fit line indicates that this
relationship is linear.
In regard to the misfit line (X = -Y), table 5 shows a significant curvature, indicating
that direct fit between the performance appraisal system and the organizational culture will
decrease more as the discrepancy between the market values of the performance appraisal
system and the organizational culture increases.
4.5.4. Hierarchy values
The regression of the hierarchy values and direct fit shows a significant result, but the
explained variance was relatively weak (R2 = 0.109, see table 4). The graph of this regression
is shown in figure 4.4.
Figure 4.4: Direct PA system – OC fit as predicted by the indirect fit of the hierarchy values of the performance appraisal system and the organizational culture. Note that PA system and OC values were centered.
In regard to the hierarchy values, table 5 only shows significant results for the fit line (X = Y).
Both the slope and curvature of this line are significant. The coefficient of the slope is
positive, indicating that the direct fit increased as the hierarchy values of the performance
Dire ct P A s y stem – OC f it
37 appraisal system and the organizational culture increased. The significant positive curvature
of the fit line indicates a non-linear relationship with upward curving. This is visible in figure
4.4.
4.5.5 Summary hypothesis 3
The regression analyses (see table 4) supported hypothesis 3 for each of the four types of
values. However, for all four value types, the fit lines did not show the straight and horizontal
shape that was expected. In all four cases, the fit line rose as the level of the values got higher.
Also, for the adhocracy, and hierarchy values, the fit line was not linear. Furthermore, direct
fit between the performance appraisal system and the organizational culture seems to suffer
more as the discrepancy between the market values of the performance appraisal system and
the organizational culture increases.
5. Discussion
The aim of this study was to investigate the effects of the fit between the values of the
performance appraisal system and the organizational culture on the satisfaction with the
performance appraisal system. In this regard, both indirect (hypothesis 1) and direct fit
(hypothesis 2) were examined, as well as the relationship between the two types of fit
(hypothesis 3). The results showed support for all three hypotheses. However, with regard to
the results some remarks need to be made. In the next section the outcomes of each of the
hypotheses will be discussed.
The first hypothesis expected indirect fit to be positively related to performance
appraisal satisfaction. Although the significant amount of explained variance for all four types
38 relationship was stronger when the levels of the values were higher rather than lower. This is
a remarkable outcome, as this was not anticipated. The first explanation that comes to mind is
that this would have been caused by a lack of differentiation in the answers of the
respondents. However, closer examination of the dataset indicates that this most likely would
not have been the case, as the range of means was broad (see Appendix 3). Thus, the
explanation has to come from a different perspective. The higher people rated the values of
the performance appraisal system, and the higher they rated the organizational culture values,
the more satisfied they were with the performance appraisal system. This phenomenon could
be a result of people wanting to reduce uncertainty (Kalliath et al., 1999). It is suggested that
people dislike ambiguity and uncertainty and will try to reduce it as much as possible. The
stronger (i.e. higher) the perceptions of values are, the more a situation is perceived as certain
(Kalliath et al., 1999). This could explain why the relationship between performance appraisal
satisfaction and the fit of the performance appraisal system with the organizational culture
was stronger when the levels of the values were higher rather than lower. Especially in current
times where alternative job opportunities are difficult to find, having a desire to reduce the
uncertainty and ambiguity of the situation seems plausible (Kalliath et al., 1999).
Furthermore, the results regarding hypothesis 1 indicated that the clan and adhocracy values
have more explanatory power than the market and hierarchy values (see table 2). This
suggests that a focus on flexibility is a better predictor of performance appraisal satisfaction
than a focus on stability. Change and adaptability are important factors in this case (Cameron
& Quinn, 2011). It could be argued that these findings are in line with the things that are
valued most by organizations nowadays. The clan values strongly emphasize the human
values, focusing on aspects such as personal development and training, and in the adhocracy
values, there is a strong emphasis on innovation and thinking outside of the box. All of which
39 is not a complete surprise that these values are represented in the performance appraisal
system as well as the organizational culture, and have relatively more explanatory power.
Moreover, the results showed that when the clan and adhocracy values were present in the
performance appraisal system rather than the organizational culture, satisfaction with the
performance appraisal system was higher. This indicates that it is even more important for a
performance appraisal system to be focused on flexibility than it is for an organizational
culture to be flexible. It is likely that the clan and adhocracy values of the performance
appraisal system are easier to observe for the employee than are the clan and adhocracy values
of the organizational culture. This may have resulted in the performance appraisal system
values having a stronger direct effect on the performance appraisal satisfaction, and thus
being more important in determining fit.
Hypothesis 2 expected direct fit to be positively related to performance appraisal
satisfaction. Results indicate that this is the case. When employees perceive the performance
appraisal system and the organizational culture to be a good fit, they will be more satisfied
with the performance appraisal system. This highlights that, in addition to the importance of
certain values of the performance appraisal system and the organizational culture in relation to
the performance appraisal satisfaction that was mentioned above, employee perception of how
well a performance appraisal systems fits within an organization is important as well. The
better the performance appraisal system is perceived to fit with the organizational culture, the
more satisfied employees are with the performance appraisal system itself.
The third and final hypothesis investigated the relationship between the indirect and
direct fit. As expected, the results indicated that indirect fit was positively related to direct fit.
However, similar to the relationship between indirect fit and performance appraisal
satisfaction, fit between the performance appraisal system and the organizational culture was
40 this may be explained by the tendency of employees to rate values higher in order to reduce
ambiguity and uncertainty (Kalliath et al., 1999). Furthermore, results indicated that direct fit
suffers the most when there is a misfit between the market values of the performance
appraisal system and the organizational culture. This may suggest that, with the exception of
market values, the values as operationalized by the Competing Values Framework do not
entirely capture the aspects which employees take into consideration when asked for fit
directly (van Vuuren et al., 2007). The fact that employees do take into account the
results-oriented aspects (i.e. the market values) when evaluating how well a performance appraisal
system fits within the organization seems plausible, as a focus on results will most likely be
something that people will always find important when evaluating a performance appraisal
system.
In all, the results offer useful insights into the relationship between the values of the
performance appraisal system and the organizational culture, and performance appraisal
satisfaction.
5.1. Practical implications
The results that were discussed above do not only offer useful insights for scholars, but also
for practitioners. Some practical implications will therefore be addressed. Organizations have
to recognize the importance of how their performance appraisal system is perceived by their
employees. The better employees perceive the system to fit within the organization, the more
satisfied they are with it. Ultimately, this will help organizations to achieve positive outcomes
(Blau, 1999; Kuvaas, 2006). Thus, it is critical that organizations do not ignore the
organizational context when implementing a performance appraisal system, and feedback
from employees plays an important role in this. Organizations should specifically pay