Linking organizational culture and change-oriented
organizational citizenship behavior
The moderating role of person-organization fit
Name: Thijs Roebroek
Study: International Business Administration, Leadership & Management Course: Master Thesis
Student Number: 10196633 Supervisor: Merlijn Venus Words: 11.409
2
Statement of originality
This document is written by Thijs Roebroek, who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
3 Abstract
Despite the importance for organizations to be increasingly adaptable in today’s complex and uncertain environment, little is known about which factors contribute to having a change-oriented organization. As employees are crucial for eliciting change within an organization, the purpose of this study is to identify antecedents of change-oriented citizenship behavior (OCB). This study focuses on the relationship between organizational culture (OC) and C-OCB and explores the role of person-organization fit (PO-fit) in moderating this relationship. The Competing Values Framework was used to distinguish between four organizational cultural types. The relationship between each cultural type and C-OCB was explored. Data was collected by conducting a survey across 162 employees across a variety of industries within the Netherlands. Results indicated that having a clan culture positively relates to C-OCB. No relationships were found between the other three cultural types and C-C-OCB. Furthermore, PO fit did not appear to moderate the relationship between organizational culture and C-OCB. However, PO-fit does seem to mediate the relationship between a clan culture and C-OCB. These findings broaden the understanding of which organizational culture types encourage COCB and thereby suggests how organizations can become more change-oriented and adaptable.
4
Table of Contents
1. Introduction ... 5
2. Literature Review ... 8
2.1 The definition of organizational citizenship behavior ... 8
2.3 Change-Oriented Citizenship Behavior ... 9
2.5 Organizational Culture ... 11
2.6 The Competing Values Framework ... 12
2.7 The relationship between organizational culture and C-OCB ... 13
2.8 PO fit as a moderator ... 15 3. Methods ... 17 3.1 Procedure ... 17 3.2 Sample ... 17 3.4 Operationalization of Variables ... 18 3.4.1 Control variables ... 18 3.4.2 Independent variables ... 18 3.4.3 Moderator ... 19 3.4.4 Dependent variable ... 19 4. Results ... 20 4.2 Correlation analyses ... 20 4.3 Hypotheses testing ... 21 4.3.1 Main effects ... 22 4.3.2 Interaction Effects ... 23 5. Discussion... 26 5.1 Summary ... 26
5.2.1 Theoretical implications of Direct effects ... 27
5.2.2 Moderation effects ... 30
5.4 Practical implications ... 32
5.6 Conclusion ... 33
References ... 34
5 1. Introduction
It has become extremely challenging to run large organizations in today’s world, which is characterized by shifts in global economic power, rapid urbanization, operational uncertainty and technological breakthroughs. Businesses are operating in increasingly complex and uncertain environments and therefore need to be able to adapt to these challenges. Consequently, organizations are forced to undergo changes almost continuously.
Employees on the work floor are more important than ever in such an environment as they are experts of the existing business practices in the organization they work at (Lawler, 1992), and are therefore better able to identify and solve problems accordingly (Morrison, 2011). Stimulating employees to initiate change, share ideas and be proactive is, therefore, an important facet of being able to adapt and ensure organizational innovation (MacKenzie, Podsakoff & Podsakoff, 2011).
One of the most discussed concepts exploring the extra-role behavior of employees is
Organizational citizenship behavior (OCB), which has been widely regarded as one of the
most important employee outcomes in organizations (Podsakoff, MacKenzie, Paine, and Bacharah, 2000). Currently, OCB is considered to be a multi-dimensional construct
consisting of discretionary behaviors that are beyond the scope of job descriptions and formal incentives (Organ, 1990). Podsakoff et al. (2000) identified seven types of OCB, which can be categorized into two dimensions: Affiliate dimensions and challenging dimensions. The first includes behaviors that reinforce the status quo by maintaining and solidifying the current practices, arrangements, and relationships. The latter are the challenging dimensions which pose a “challenge” to the status quo by “voluntary acts of creativity and innovation designed to improve one’s task or the organization’s performance” (Podsakoff et al., 2000, p. 524). The challenging dimension, particularly important in today’s world as it is assumed to be initiating change in an organization, is examined in this research as change-oriented
organizational citizenship behavior (C-OCB).
Although OCB has been widely studied in the past decades, most studies have
concentrated on the affiliate dimensions of OCB. Despite their importance for organizational change, limited research has been conducted on C-OCB (Ashworth, 2012; Choi, 2007; Datta, 2012. Existing literature has mainly focused on individual dispositional variables such as personality, motivational states, self-efficacy and coping (Bettencourt, 2004; Frese, Fay, Hilburger, Leng, & Tag, 1997; Lepine & Van Dyne, 1998). More recently, a few of the
6 situational variables have been studied such as leadership (Van Dyne, Kamdar, & Joireman, 2008; Sagnak, 2016) and social context (Chiaburu, Peng, Oh, Banks & Lomeli, 2013).
One promising variable for predicting C-OCB is organizational culture. This is because an organizational culture provides employees with a reference point for evaluating whether their behaviors are acceptable and appropriate according to organizational norms and expectations. This framework of organizational culture thereby significantly influences employee behavior (Atchison, 2002; Cooke & Rousseau, 1988) and thus can be a powerful instrument to induce change-oriented behaviors in organizations. On top of this, Aguirre et al. (2013) reported that amongst a sample of 2,200 top executives around the world, 84%
claimed that organizational culture is the most important factor in the process of change initiatives.
Only limited research has been done on the effect of organizational culture on C-OCB. Choi (2007) found a positive relationship between an innovative organizational culture and C-OCB but emphasized the need for further theoretical and empirical development in the area. Patterson et al. (2005) also concur, suggesting for more research on the effects of specific dimensions of organizational culture on a broad range of outcomes. This study aims to extend the line of research by Choi (2007) by investigating the relationship between organizational culture and C-OCB using the Organizational Culture Assessment Inventory
(OCAI).
Another important variable that is known to have a significant impact on employee behavior is Person – Organization Fit (PO fit), which is defined as “the compatibility between people and the organizations in which they work” (Kristof, 1996, p.1). Previous research found that PO fit affects employee’s behaviors (Hoffman & Woehr, 2006) and is a powerful antecedent of OCB (Farzaneh, Farashah & Kazemi, 2014; de Lara, 2008; Vilela, Gonzalez & Ferrin, 2008). Therefore, it represents an important variable worth considering when studying organizational culture. Employees arguably respond differently to different organizational cultures, and therefore PO fit might influence he relationship between organizational culture and C-OCB. It is predicted that PO fit influences this relationship in such a way that if an employee believes his or her values are congruent with those of the organization, the effect of organizational culture on C-OCB is likely to be stronger.
Consequently, for the purpose of this study, PO fit is considered to be a possible moderator of the relationship between organizational culture and C-OCB in this study.
In this study it is hypothesized that both the clan and ad hoc culture have a positive relationship with C-OCB. On the other hand, it is expected that the internal process culture
7 will have a negative relationship with C-OCB. It is predicted that PO fit will moderate this relationship in such a way that a high PO fit will make these relationships stronger, and a low PO fit will make them weaker. The research model of this study can be found in figure 1.
A deeper understanding of the relationship between organizational culture and C-OCB, and the role of PO fit in this relationship, may provide organizations a better insight into what kinds of working environments influence employees as drivers for organizational change. This knowledge can be used to structure the organization in such a way that it maximizes the C-OCB of its employees and therefore creates an organization that is flexible and capable of continuous change. Furthermore, exploring the role of PO fit can have significant consequences for the selection, attrition, and retention of the right employees.
The next chapter reviews the relevant literature of the theoretical concepts discussed above. Subsequently, the method chapter will state the research procedure, the sample, and the operationalization of the variables used in this study. The following chapter presents the findings, followed by the discussion which discusses the theoretical and practical
implications of the results. The final chapter is devoted to a conclusion of the research, summarizing the main findings, discussing the limitations and shortcomings of the analysis, and making suggestions for future research.
Figure 1. Research model.
8 2 Literature Review
This chapter reviews the literature to define the key terms and identify the relevant variables which will be analyzed in the research. An in-depth literature review which defines and discusses the relevant terms of this study is conducted. This literature review elaborates on the terms “organizational citizenship behavior”, “change-oriented citizenship behavior”, “organizational culture”, “the competing values framework” and “PO fit”. Lastly, the hypotheses of this study are stated, together with the theoretical support for each of these hypotheses.
2.1 The definition of organizational citizenship behavior
Daniel Katz (1964) was the first to distinguish between employee in-role behaviors and “spontaneous and innovative behaviors” (p. 132) that are necessary to meet organizational goals. Smith, Organ, and Near (1983) expanded on Katz’ work by referring to it as
“organizational citizenship behaviors” (OCB) (p.1). They constructed two dimensions of OCB, namely altruism and general compliance. A few years later, Organ (1988) defined OCB as:
Individual behavior that is discretionary, not directly or explicitly recognized by the formal reward system, and that in the aggregate promotes the effective functioning of the organization. By discretionary, we mean that the behavior is not an enforceable requirement of the role or the job description, that is, the clearly specifiable terms of the person’s employment contract with the organization; the behavior is rather a matter of personal choice, such that its omission is not generally understood as punishable (p.3).
Academic attention to OCB has been growing since the 1980s when the term was first introduced in the academic literature. Since then nearly 200 articles that are related to OCB have been published (Podsakoff et al., 2000). However, the literature on OCB is fragmented since there is no consensus and consistency regarding the terminology and taxonomy of OCB. Other constructs that are related to Organ’s (1988) definition are “prosocial organizational behavior” (George Van Dyne, Kamdar, & Joireman, 2008; O’Reilly & Chatman, 1986), “employee voice” (LePine & Van Dyne, 1998), “organizational spontaneity” (George & Jones, 1997), “social participation” (Vandewalle, Van Dyne & Kostova, 1995) and “contextual performance” (Borman & Motowidlo, 1993).
Organ’s original definition (1988) has also been criticized by multiple sources with the “discretionary” element of the definition particularly being noted. For example, Morrison (1994) found that 18 out of the 20 items of Podsakoff et al.’s OCB scale was described by
9 respondent’s as in-role and thus seen part of the job. He concluded, “OCB is ill-defined and varies from one employee to the next” (p. 1561). Organ (1997) later admitted this problem and attributed it to the “fuzziness of the concepts role and job” (p.88). Another criticism regarded the mentioning of the formal reward system (Organ, 1997), as multiple studies found that OCB behavior was just as likely to receive a monetary reward as in-role performance (MacKenzie, Podsakoff, & Fetter, 1991; Podsakoff & MacKenzie, 1994;
Werner, 1994). Organ recognized the above limitations and therefore redefined his definition of OCB as “performance that supports the social and psychological environment in which task performance takes place” (1997, p. 95).
One important outcome of OCB is that it stimulates the adequate functioning of an organization” (Organ, 1983). Research regarding antecedents has mainly focused on
individual, task and organizational characteristics (Podsakoff et al., 2000). Earlier research mainly studied individual characteristics. A meta-analysis by Organ and Ryan (1995) found that an employee’s job attitudes, such as job satisfaction and commitment, predict OCB better than dispositional characteristics. Furthermore, turnover intention is found to be negatively related to OCB (Chen, Hui & Sego, 1998). Subsequent research on context characteristics has mainly been on the topic of leadership. Transformational leadership is found to be a relatively stable predictor of OCB (Wang, Law, Hackett, Wang & Chen, 2005).
2.3 Change-Oriented Citizenship Behavior
A review by Podsakoff et al. (2000) revealed that almost 30 OCB-related constructs have been identified and studied. Their analysis suggested that many of these constructs overlap. Therefore, these constructs could be organized into seven dimensions: Helping Behavior, Sportsmanship, Organizational Loyalty, Organizational Compliance, Individual Initiative, Civic Virtue, and Self-Development.
Van Dyne and LePine (1998) found that these dimensions can be split up in ‘affiliation-oriented behaviors’ and ‘challenge-‘affiliation-oriented behaviors.’ Affiliative behaviors are ‘affiliation-oriented towards maintaining current relationships and arrangements. Challenge-oriented behaviors, in contrast, are future-oriented and are concerned with making constructive changes in the work environment. The literature has shown that these two types of OCB behaviors are factorially distinguishable and each type accounts for a unique variance of overall performance
(Whiting, Podsakoff & Pierce, 2008). Six out of the seven dimensions construed by
10 be challenge-oriented. Morrison and Phelps (1999) argued that affiliate behaviors “are not sufficient for ensuring the continued viability of an organization and that organizations also need employees who are willing to challenge the present state of operations to bring about constructive change” (p.403).
Since then, scholars have drastically increased their interest in the challenge-oriented type of OCB. Different constructs have been used to study challenge-oriented OCB such as voice (Van Dyne & Lepine, 1998), innovative behavior (West & Farr, 1990), taking charge (Morrison & Phelps, 1999), change-oriented behavior (Marinova, Peng, Lorinkova, Van Dyne & Chiaburu, 2015), personal initiative (Frese & Fay, 2001), proactive work behavior (Parker, Bindl & Strauss, 2010), creative performance (Zhou & George, 2001) and change-oriented Citizenship Behavior (C-OCB) (Bettencourt, 2004). All of these concepts represent change-oriented behavior that is future-focused and self-initiated. A minor difference is that some of these constructs only focus on idea suggestion, such as voice, while others also focus on the implementation of these ideas.
C-OCB covers both idea suggestion and its implementation (Seppala, Lipponen, Bardi & Pirttilla-Backman, 2012). Since this seems more powerful than solely expressing ideas, this research focuses on C-OCB. Choi (2007) defined change-oriented OCB as
“Constructive efforts by individuals to identify and implement changes, with respect to work methods, policies, procedures to improve the situation and performance” (p.28). By doing so the current status quo is challenged, in contrast to affiliative-OCBs which intend to reinforce the status quo.
Lopez-Dominguez, Enache, Sallan & Simo (2013) note that despite being drivers for organizational change, only a few scholars have paid attention to C-OCB to date. Lepine and Van Dyne (1998) found several employee attitudes such as job satisfaction and commitment to be related to C-OCB. This was later confirmed by Parker, Bindl & Strauss (2010) as well as by Farh and Farh (2012). Parker, Williams and Turner (2006) found that co-worker support and supportive supervision are also positively related to change-oriented OCB. Most other research on contextual variables has been on leadership (Bettencourt, 2004; Choi, 2007; Lopez-Dominguez et al., 2013; Vigoda-Gadot & Beeri, 2011). For example, transformational leadership was found to promote C-OCB among employees as it stimulates employees to perform above their normal level and improves commitment, which in turn stimulates C-OCB (Bettencourt, 2004; Lopez-Dominguez et al., 2013).
One promising contextual variable that has received little attention is organizational culture. Organizational culture is widely assumed to influence employee behavior. Hence it
11 provides employees a reference point against which the appropriateness of behaviors may be evaluated (Atchison, 2002). Therefore, it might be a powerful tool to induce much-wanted employee behaviors such as C-OCB. For these reasons, organizational culture is a promising variable to study. Little research has been done on the relationship between organizational culture and C-OCB. Only Choi (2007) found a positive correlation with innovative climate and C-OCB. However, there is room for more theoretical and empirical development. Therefore, this study aims to expand Choi’s research by exploring the relationship between different types of organizational cultures and C-OCB.
2.5 Organizational Culture
The term "organizational culture" (OC) has its roots in cultural anthropology but is now a prevalent topic in the field of organizational behavior. More than 4000 articles have been published on the subject since 1980 (Hartnell, Ou, & Kinicki, 2011). One of the first definitions of OC came from Forehand and Vol Gilmer (1964), who argued that OC is a combination of characteristics that distinguish organizations from one another. Tichy (1982) described OC as the normative glue that keeps organizations together.
There are so many overlapping and competing definitions of that Brown (1995) called this “an embarrassment of definitional riches” (p. 5). However, there are certain
commonalities between the definitions. Researchers seem to agree that OC is shared among employees, exists at multiple levels of the organization, influences employees' attitudes and behaviors, and consists of collective beliefs, assumptions, and values (Ostroff, Kinicki, & Tamkins, 2003).
Schein (1985) construed a comprehensive definition that incorporates these dimensions into a single definition of OC:
a pattern of shared basic assumptions that the group learned as it solved its problems of external adaptation and internal integration, that has worked well enough to be considered valid and, therefore, to be taught to new members as the correct way to perceive, think, and feel in relation to those problems (p.19).
In subsequent studies Schein (1990) later suggests that the concept should be broken down and distinguished into three cultural levels; basic assumptions, espoused values, and artifacts. Basic assumptions are the deepest layer and refer to the core, or essence of the organization. These are unconscious shared beliefs. Espoused values are the middle layer and are the fundamental principles, morals, and values of the organization. They are the beliefs
12 upon which a company is built. The top layer of culture, referred to as artifacts, is the visible manifestations and the tangible elements within the organization.
2.6 The Competing Values Framework
Many methods to measure OC have been developed by scholars over the years, hence resulting in a vast amount of scales and models (Yu & Wu, 2009). This study uses the
Competing Values Framework (CVF) by Cameron and Quinn (2006), which integrates the
major dimensions of OC into one single framework. (Patterson et al., 2005). Ralston, Terpstra-Tong, Terpstra, Wang and Egri (2006) found that it covers all eight relevant
dimensions of OC identified in a review study. The CVF is the most extensively studied and used model in quantitative research on OC (Kwan & Walker, 2004; Ostroff et al., 2003). The CVF has been used in more than 10,000 organizations worldwide (Cameron& Quinn, 2006). Furthermore, the reliability and validity of this framework has been empirically verified through multiple techniques such as multidimensional scaling and structural equation modeling (Howard, 1998; Quinn & Spreitzer, 1991; Yu & Wu, 2009).
The framework consists of two dimensions. The first dimension is the organizational focus, which ranges from an internal micro-orientation to an external macro-orientation. The second dimension is related to organizational structure, which ranges from an emphasis on stability to a focus on flexibility. These two dimensions lead to four distinct OC types that will be discussed below. The four OC types are differentiated based on the following six characteristics: dominant organizational characteristics, leadership style, management of employees, organizational glue, strategic emphasis and criteria for success.
The clan culture has a flexible structure with an internal focus. It has similarities with a
family-like organization, where there is a high concern for human relations. Its core values are trust and support which result in a high level of commitment and cohesion. Behaviors associated with this type of culture are teamwork, participation and open communication (Cameron & Quinn, 2006). Its leadership style is facilitative.
The ad hoc culture has a flexible structure with an external focus. It symbolizes a
dynamic and highly entrepreneurial creative environment where employees have significant autonomy. Behaviors that belong to this type of culture are risk-taking, creativity, and adaptability. Leaders are visionary and innovative. The metrics for measuring the
effectiveness of this type of culture includes resource acquisition and the development of new markets (Hartnell et al., 2011).
13
The internal process culture has a stable structure with an internal focus. It is driven
by control, formalization, and standardization. Cameron and Quinn (2006) compared it to a bureaucracy like culture. Policies, well-defined roles, and regulations govern the behavior of its employees. Typical behaviors are conformity and predictability. Leaders are coordinating and controlling.
The market culture has a flexible structure with an external focus. It emphasizes
productivity, performance, and achievement. Key characteristics are result-orientated and competitiveness. Behaviors associated with this type of culture are planning, task focus, and articulation of clear goals. Leaders in this kind of culture tend to be hands-on, result-oriented and directive.
2.7 The relationship between organizational culture and C-OCB
The different cultural types promote different kinds of employee behaviors. Therefore it stands to reason that there may be variances in how well cultural types predict C-OCB. As previously mentioned the clan culture possesses a high level of concern for
employees. Support is, therefore, one of the core dimensions of this kind of culture. C-OCB challenges the status quo. This can be seen as a risk since this can contribute to tension between employees or working units. Support from various resources can therefore be critical, as without it employees are not willing to take such risks. Chiaburu et al. (2013) found a positive relationship between the social context (leader support, coworker support, and organizational support) and C-OCB. Choi (2007) and Parker et al. (2006) also found a positive relationship between leadership support and C-OCB. The clan culture has a similar social context, which in turn can induce C-OCB. Trust is another important facet of the clan culture. Previous research found a positive relationship between trust and innovative work behavior (Clegg, Unsworth, Epitrpaki & Parker, 2002; Parker et al., 2006). If employees experience trust, they are likely to gain a sense of self-efficacy. Self-efficacy is important when engaging in C-OCB. Such behaviors entail individual risks and are more likely to be elicited if employees are confident that their actions will be successful (Parker et al., 2006). Trust signals that mistakes are acceptable and seen as learning experiences. These factors motivate employees to try things beyond their core role.
Since trust and support are important antecedents of C-OCB and are both core characteristics of a clan culture, it is predicted that:
14
Hypothesis 1a: A clan culture has a positive relationship with change-oriented citizenship behaviors.
C-OCB is a self-initiated proactive behavior. One important requisite for an OC that wants to promote C-OCB, is to encourage self-initiation. The internal process culture is driven by control, formalization, routinization, and centralization. This culture has clear role descriptions, which limits the opportunities for employees to engage in C-OCB. A high degree of formalization incorporates many specific rules that need to be followed (Perrow, 1967), which leaves less room for C-OCB. Subsequently, routinization prevents initiation due to repetitiveness (Ohly, Sonnentag, & Pluntke, 2006). Both routinization and formalization were found to be negatively correlated with C-OCB (Marinova et al., 2015), because these characteristics “prevent experimentation with new ideas and opportunities to learn new skills” (Marinova et al., 2005, p.108). It is therefore predicted that:
Hypothesis 1b: The internal process culture has a negative relationship with change-oriented citizenship behaviors.
The ad hoc culture is characterized by autonomy, which is found to be a major determinant of change-oriented behaviors such as ‘employee voice’ (Van Dyne & Lepine, 1998) and ‘personal initiative’ (Frese et al., 1997). It is found that when employees view their job as autonomous, they are likely to go beyond the role descriptions and elicit C-OCB (Binnewies & Gromer, 2012). Likewise, multiple studies found a positive relationship between job autonomy and C-OCB (Marinova et al., 2015; Parker et al., 2006). Autonomy gives employees the freedom in their respective role to elicit this type of voluntary behavior. Cunningham et al. (2002) argue that job autonomy gives an individual the feeling it has control over one’s situation. This decreases the sense of threat an employee feels towards change, thus making the employee more receptive to change (Cunningham et al., 2002).
Leaders in the ad hoc culture have a strong vision, which is a core component of a firm’s capability to successfully implement change (Zaccaro & Banks, 2004). Parker, Bindl, and Strauss (2010) build on this by stating: “Vision provides a discrepancy between the ideal situation and the current situation, thereby providing a motivational force for proactive action” (p.845). Another effect of a strong vision is that it generates intellectual stimulation and inspiration among employees, which then promotes change-oriented behavior
15 “transformational leadership”(Belasen & Frank, 2008), which is also positively related to C-OCB (Burris, Detert, & Chiaburu, 2008; Janssen & Van Yperen, 2004; Strauss, Griffin, & Rafferty, 2009). A positive relationship exists between the flexible culture types (ad hoc and market) and innovative work behavior. Lastly, an innovative climate, closely resembling the ad hoc culture, is positively related to C-OCB (Choi, 2007). Based on the above arguments, it is predicted that:
Hypothesis 1c: The ad hoc culture has a positive relationship with change-oriented citizenship behavior.
The market culture emphasizes performance, task-orientation, competitiveness, and individualism. Employees are motivated by clear goals and contingent rewards in this type of culture. People are expected to take care of themselves, and therefore it is rather
individualistic. Individualism has been found to be negatively related to OCB (Moorman & Blakely, 1995). The leadership style in the market culture is directive and goal-oriented. If employees get directed what to do, they might be less likely to show self-initiated extra-role behaviors. Based on the above arguments, one could propose a negative relationship between the market culture and C-OCB. However, the focus on performance might also positively stimulate C-OCB. Employees are expected to display superb performance and are rewarded well for it. This might stimulate employees to go the extra mile by showing additional role behaviors such as C-OCB. Furthermore, the market culture is externally oriented, always trying to improve and change for the better to keep up with competitors. Therefore, the market culture might stimulate change-oriented behaviors and may thus improve the competitiveness and performance of the company.
In short, one may expect that a relationship between the market culture and C-OCB exists, but no directional hypothesis could be asserted based on the literature. Therefore it is assumed no relationship will be found in this study.
2.8 PO fit as a moderator
Person - Organisation fit (PO fit) is defined as “the compatibility between people and
the organizations in which they work” (Kristof, 1996, p.1). The exact nature of this
compatibility is subject to much debate. PO fit has been defined in terms of goal congruence, value congruence, needs-supplies fit, and demands-abilities fit (Kristof, 1996). This study
16 uses value congruence since it is the most widely assessed dimension of PO fit. Cable and DeRue (2002) narrowed Kristof’s definition by describing PO fit as an employee’s subjective beliefs about how congruent his or her values are with that of the organizational culture. The value congruence between employees and their organization affects employee behavior. PO fit has been linked to attitudinal and behavioral employee outcomes. Hoffman and Woehr (2006) found that PO fit is related to job performance, OCB, and turnover. Two meta-analyses revealed that PO fit has a strong relationship with both job satisfaction and organizational commitment and a moderate relationship with intention to quit (Kristof-Brown, Zimmerman & Johnson, 2005; Verquer, Beehr & Wagner, 2003). Other studies have also confirmed the positive relationship between PO fit and OCB (de Lara, 2007; Farzaneh et al.,2014; Vilela et al., 2008).
How employees react to the type of organizational culture might depend on how well their values match with those of the organization. If employees have a high PO fit and thus share the same values as the organization, it may then be concluded that they have a good understanding of the wants and needs of the organization (Erdogan & Bauer, 2005) and are therefore more likely to agree with the goals of the organization. Organizational culture can be seen as a value framework drawing the boundaries for the appropriate types of behaviors. The signals about appropriate behaviors that the organization sends might be better
understood and embraced by employees with a high PO fit. Therefore, the effects and
outcomes of organizational culture on an employee’s behavior are likely to be stronger in the cases of high PO fit.
On the other hand, employees with a low PO fit are not likely to share the same values as the organization and therefore less likely to understand the needs and wants of the organization and embrace its culture. Subsequently, if an employee has a low PO fit, the effects of the type of culture on his or her behavior is likely to be weaker. Thus, it is
hypothesized that PO fit affects the relationship between organizational culture and C-OCB in such a way that the relationship between the type of organizational culture and C-OCB will be stronger for employees with a high PO fit and weaker for employees with a low PO fit. This leads to the following hypotheses:
Hypothesis 2a: PO fit will moderate the relationship between the clan culture and C-OCB in such a way that a high PO fit will make the relationship stronger and a low PO fit will make
17
Hypothesis 2b: PO fit will moderate the relationship between the ad hoc culture and C-OCB in such a way that a high PO fit will make the relationship stronger and a low PO fit will
make the relationship weaker.
Hypothesis 3b: PO fit will moderate the relationship between the internal process culture and C-OCB in such a way that a high PO fit will make the relationship stronger and a low PO fit
will make the relationship weaker.
3 Methods
This chapter provides information on the research approach and methodology used for this study. First, the procedure of the study will be stated, followed by an evaluation of the study sample. Lastly, the operationalization of the variables will be discussed along with their respective reliabilities.
3.1 Procedure
The data for this research was collected through an online survey and analyzed with the statistical software SPSS. Qualtrics, an online survey tool, was used to develop and distribute the survey. The data was collected in a 3-week period. Non-probability sampling techniques were used to recruit participants, such as convenience sampling through email and snowballing. The surveys were distributed through Qualtrics, whereby the participants
received an email with a personal link to the survey. Several reminders were sent through Qualtrics, which is likely a contributing factor to the high response rate of (83%). It took respondents an average of 7 minutes to complete the survey. The study had aimed to collect information from at least 100 people but was successfully able to obtain data from 162 participants. The survey questionnaire can be found in the Appendix.
3.2 Sample
Out of the 162 participants, 25 cases were deleted because of missing data. This left 137 respondents, of which 84 are male (61.3%), and 53 are female (38.7%). The age ranged between 18 and 58 with an average age of 28 years (SD=10.37). The sample is highly educated, with 75.9% (N=104) having a university degree (WO) and 16.8% (N=23) a higher education (HBO) degree. More than a third of the respondents, 37.2% (N=51) work in the financial and business sector, followed by a group of 23.4% (N=32) who indicated that they
18 work in an industry other than the options provided in the survey. The sample has an even distribution of full-time (50.4%) and part-time (49.6%) employees. The average tenure is 3.47 years (SD=5.39) with a range from 0.2 to 26.5 years.
3.4 Operationalization of Variables
Multiple demographic variables such as age, gender, education, and type of industry were collected to sketch a clear picture of the sample. All scales have been previously validated and use a 7-point Likert scale that ranges from (1) “completely disagree” to (7) “completely agree”. Two out of three scales have been translated from English to Dutch, using a two-way back-and-forth translation technique. The Organizational Culture Assessment Inventory (OCAI) was already available in Dutch and therefore did not need translation.
The internal consistency is measured using the Cronbach’s alpha. Kline (2013) states a score between .5 and .6 is acceptable, and items that score above .7 are considered to have high reliability. Subsequently, the item-total correlation is checked for each item, to see if each item correlates with the overall score from the scale.
3.4.1 Control variables
Two additional variables have been included to control for systematic biases. Tenure in years has been included since it may be assumed that employees who have been working at the same organization for a long time may tend to have a higher PO fit. For this variable, the assumption of linearity does not hold up since an extra year is not likely to increase or decrease PO fit significantly. Therefore, a dummy variable was introduced to make a distinction between employees who have worked for the same company for a short (0=less than two years), versus a long period (1=more than two years). Furthermore, Contract Hours, coded 0 if part-time (less than 36 hours a week) and 1 if full-time (more than 36 hours a week), is also used as another control variable in the model.
3.4.2 Independent variables
Organizational Culture was measured using the Organizational Culture Assessment
Inventory (OCAI). This is the most widely used and validated scale of organizational culture
currently available based on the competing values framework. Several researchers have provided evidence for its reliability and validity (Quinn & Spreitzer, 1991; Yeung,
19 and therefore has six items per cultural type. The six dimensions are ‘dominant
characteristics’, ‘organizational leadership’, ‘management of employees’, ‘organizational glue’, ‘strategic emphasis’ and ‘criteria of success’. To keep the survey concise and to the point, the scale was shortened to 16-items based on a previously conducted factor analysis by Suppiah and Sandhu (2016), whereby the items for the dimensions ‘strategic emphasis’ and ‘criteria for success’ were deleted from the scale.
The cultural types that were assessed in the study are as follows: Clan, ad hoc,
internal process and market. An example statement used in the survey to assess the clan
culture is: “The leadership in the organization is generally considered to exemplify
mentoring, facilitating, or nurturing” (‘leadership’ dimension). An example statement to
assess the ad hoc culture is: “The organization is a dynamic and entrepreneurial place.
People are willing to stick their necks out and take risks” (‘dominant characteristics’
dimension). A statement for the market culture is: “The management style in the organization
is characterized by hard-driving competitiveness, high demands, and achievement”
(‘management of employees’ dimension). Lastly, an example for the internal process culture is: “The glue that holds the organization together is formal rules and policies. Maintaining a
smooth-running organization is important” (‘organizational glue’ dimension).
Reliabilities of the culture scales were satisfactory. All items in the scales had all item-total correlations >.30. The Cronbach’s Alpha values for the clan, ad hoc, internal process, and the market scales were .89, .86, .79 and .89, respectively.
3.4.3 Moderator variable
Based on previous academic literature, PO fit is studied as a potentially suitable moderator for the relationship between OC and C-OCB. Person - Organization fit is defined as “the employees’ subjective perceptions of the fit between their values and the values of the organization” (Cable & Judge, 1996, p.299). It was measured by a 3-item scale created by Cable and Judge (1996), which is one of the most widely used scales in the literature and has been extensively validated. The P-O fit scale had Cronbach’s alpha of α=.78, with each corrected item-total correlation being above .30. This means the scale was appropriate to use for this study.
3.4.4 Dependent variable
Change-oriented citizenship behavior was selected as the dependent variable of this research study. Change-oriented Citizenship Behavior is defined as: “Constructive efforts by
20 individuals to identify and implement changes, with respect to work methods, policies,
procedures to improve the situation and performance” (Choi, 2007, p.27). C-OCB was measured by Choi’s (2007) 4-item scale, adopted from Scott and Bruce (1994) and Morrison and Phelps (1999). The reliability of the scale is high, with a Cronbach’s Alpha of .86. The reliability analyses, as mentioned above, indicate that all the scales utilized in this study have a good reliability value, suggesting that the data obtained from these scales can confidently be analyzed to draw predictive conclusions. Furthermore, all items within each scale provided sufficient inter-item reliability, and therefore no items had to be deleted.
4 Results
This chapter of the research presents the results of the data analysis and the findings of the research study. First, SPSS was used to create a series of correlations between factors and then used to test the hypotheses through a hierarchical multiple regression model.
Before moving on to the heart of the data analysis, some preliminary analyses were conducted. Out of the 162 respondents, 25 cases were deleted for the reason of missing data, which left 135 cases for the statistical analyses. One item was counter-indicative and had to be recoded. The data was tested for skewness, kurtosis, normality, and outliers. No
noteworthy results were found for these tests. Prior to the data analyses, all scale scores were standardized and the interaction terms computed, in order to be able to conduct a moderation analysis.
4.2 Correlation analyses
An overview of the descriptives, correlations and scale reliabilities are presented in Table 1. The mean scores for all the scales were relatively high, in particular for PO fit (M= 5.25, SD=1.32). Regarding the control variables, only Tenure is significantly correlated to
C-OCB (r(135) = .19, p < .05). Some interesting observations are the inter-item correlations of
the cultural types since some of these correlate with each other. Particularly Clan and Ad Hoc (r(135)=.70, p < .05) are highly correlated, but so are Market and Internal Process
(r(135)=.33, p< .05), Market and Ad Hoc (r(135)=.29, p <.05), and Clan and Internal Process (r(135)=.18, p<.05). Lastly, Clan is correlated to all cultural types, C-OCB, and PO fit. Both Clan (r(135)=.41, p<.01) and Ad Hoc (r(135).59, p<.01) are correlated with PO
fit. This suggests that respondents who experience a Clan or Ad Hoc in their organization,
21
p<.01) and Ad Hoc (r(135)=.29, p<.01) are correlated with C-OCB. This suggests these
cultural types promote more C-OCB than other cultures. Note though, the direction of this relationship cannot be conceived since causality cannot be confirmed. Lastly, a positive correlation between C-OCB and PO fit also exists (r(135)=.48, p <.05). This suggests employees who experience a high PO fit, are more likely to engage in C-OCB. Another possibility is that employees who engage in C-OCB are more likely to perceive a fit with their organization.
Note: Reliabilities are reported along the diagonal
** Correlation is significant at the 0.01 level (2-tailed) * Correlation is significant at the 0.05 level (2-tailed)
4.3 Hypotheses testing
A stepwise hierarchical multiple regression model was constructed to investigate the relationship between organizational culture and C-OCB. C-OCB is the dependent variable,
Clan, Internal Process, Ad Hoc, and Market serve as the main independent variables, and
Tenure and Contract Hours as the control variables in the regression model. A moderation analysis was conducted to investigate if PO fit acts as a moderator in the relationship between the four organizational cultures and C-OCB. Table 2 below presents the findings.
Table 1. Descriptive Statistics: Means, Standard Deviations, inter-scale correlations and
Reliabilities (N=135) Variables M SD 1 2 3 4 5 6 7 8 1. Tenure .31 .466 - 2. Contract Hours .50 .502 .1 - 3. Clan 4.89 1.35 -.03 0.04 (.88) 4. Internal Process 4.16 1.25 -.07 .05 .18* (.79) 5. Ad Hoc 4.42 1.34 -.19* -.08 .71** .02 (.86) 6. Market 4.20 1.48 -.15 .01 .11 .33* .29* (.89) 7. PO Fit 5.25 1.32 .00 -.03 .67** -.03 .59** .10 (.78) 8. C-OCB 4.74 1.16 .19* .14 .41** -.02 .29** -.08 .48** (.86)
22 4.3.1 Main effects
The first model in Table 2 assesses whether the control variables, Tenure and Contract
Hours, are independently related to C-OCB. The findings indicate that both Tenure and Contract Hours are statistically significant and explain 5% of the variance in C-OCB (R2=.05,
F(2.13 =3.5, p <.05). In this model only Tenure showed a significant correlation with C-OCB
(β=.18, p < .05).
The second model is constructed to test Hypothesis 1a. In step 1, the control variables were entered, followed by the variable Clan in step 2. After entering Clan, the model
explained 22% of the variance (R2= .22, F(3.13)=12.59, p < .01). It explained 17% additional variance after controlling for Tenure and Contract Hours (R2 change=.17, F(1.13) =29.13, p < .01). So, although Tenure explains a minor variance in this model, Clan by far has the most predictive power. In this model two out of three variables were significant, with Clan
recording a higher Beta value (β=.41, p <.01) than Tenure (β =.19, p <.05). In other words, a tenure of more than 2 years in the same organization is positively correlated with C-OCB. Due to its reported significance, it can be stated that Clan culture predicts C-OCB. This provides sufficient evidence to support a confirmation of Hypothesis a1, and thus the hypothesis is accepted.
The third model tests Hypothesis 1b. The control variables were added first, followed by Internal Process. The model as whole was not significant (R2 change = .00, F(3,13)=2.37,
p > .05). Internal Process did not have a significant Beta (β =-.01, p > .5). However, Tenure
did have a significant effect in this model (β=.18, p < .05). In summary, the results of the third model fail to provide conclusive evidence in support of Hypothesis 1b. Therefore, the hypothesis cannot be accepted
Model 4 tests the impact of Ad Hoc on C-OCB. The control variables were entered in step 1, followed by the variable Ad Hoc in step 2. The model explained 17% of the variance in C-OCB (R2=.17; F(3,13)=9.13; p < .01). Adding Ad Hoc explained 12% additional
variance (R2 change=.12; F(1,13)=19.23; p < .01). In this model both Tenure (β=.25, p < .01) and Ad Hoc (β=.35, p < .01) were statistically significant. Conclusively, introducing Ad Hoc to the model significantly contributed to predicting C-OCB, thereby providing evidence to support Hypothesis 1c. As such, H1c is accepted.
Model 5 explores the relationship between a Market culture and C-OCB. Following the control variables in step 1, Market is added to the model in step 2. The model explained 5.3% of the variance but was not significant (R2=.05, F(3,13)=2.5, p > .05). Market was found to be non-significant (β=-.05, p > .05), whereas Tenure emerges as the only significant
23 variable in the fifth model (β=.17, p < .05). These results are in line with the expectation that no relationship was expected to be found between Market and C-OCB.
Model 6 includes all four cultures and control variables. The culture variables
Clan, Internal Process, Ad Hoc, and Market were added following the entry of the control
variables. The model explained 24% of the variance (R2=.24, F(6,13)=6.92, p <.01). It
explained an additional 19% variance on top of the control variables (R2 change= .19, F(4,13) =8.2). Unexpectedly, only Tenure (β=.19, p <.05) and Clan (β=.333, p < .01) were
statistically significant, while Ad Hoc was not a significant predictor of C-OCB (β=.15, p > .05). Therefore, one may suggest that the previously observed effect of Ad Hoc on C-OCB may actually be a “shadow effect,” pointing to a spurious relationship. These results indicate that Hypothesis 1c is not supported by the data, and thus the hypothesis is rejected.
4.3.2 Interaction Effects
PO fit was considered as a moderator in the relationship between the cultural types and C-OCB. The following section describes the results of the moderation analysis, which tests hypotheses 2a, 2b, and 2c. To that aim, four additional models were constructed, which are presented in Table 2. Each contains a variable that interacts PO fit with one of the four culture variables.
Model 7 tests Hypothesis 2a. It introduces a new variable that interacts Clan and PO fit. It first adds the control variables, Clan and PO fit, and then the Clan*PO fit interaction. The model did not yield any a significant increase in the variance (R2 change=.02, F(1,13)= 3.34). The interaction term was also non-significant in the model (β=-.16, p >.05). Thus, PO fit does not appear to be a significant moderator in the relationship. The results of this model do not support Hypothesis 2a. The model also shows that both Tenure (β=.17, p <.05) and PO fit (β=.33, p < .01) were significant explanatory variables of C-OCB. Interestingly, once PO fit
is added to the model, Clan is not a significant predictor of C-OCB anymore(β=.12, p > .05).
Hypothesis 2b predicted that PO fit would moderate the relationship between Clan and
C-OCB in such a way that a high PO fit would make the relationship stronger whereas a low PO fit would make it weaker. Model 8 was constructed to this hypothesis. When Internal Process and the interaction term between Internal Process and PO fit were added to the
model there was no significant increase in the explained variance (R2 change=.02, F(1,13)= 3.6, p > .05) and the interaction term was not statistically significant (β=-.15, p > .05). The results suggest that PO fit does not act as a moderator in the relationship between Internal
24 of Tenure (β= .17, p < .05) and PO fit (β=.46, p < .001) to C-OCB, while Internal Process was found to be not a significant indicator (β= .03, p >.05).
Model 9 was constructed to investigate Hypothesis 2b. Tenure, Contract Hours, Ad
Hoc and the interaction term between Ad Hoc and PO fit were regressed against the variable
C-OCB. The model did not provide any additional explanatory power (R2 change = .00
F(1,13)=.25; p > .05) and the interaction term yielded no significant results (β=.04, p >.05).
The coefficients for Tenure (β=.19, p <.05) and PO fit (β=.51, p<.001) were positively significant but the coefficients for Ad Hoc (β=.10, p >.05) and Contract Hours (β=.14, p > .05) were not.
Next, model 10 was developed to test the null hypothesis regarding the market culture. In this model, the control variables, Market, PO fit, and the interaction term between the two were taken as the independent variables. Similar to the previous three models, no extra variance was explained by this model (R2 change= .29; F(1,13)=1.1; p >.05) and the
interaction term was non-significant (β=.08, p > .05). Furthermore, Market (β=-.09, p > .05) and Contract Hours (β= .14, p >.05) were not found to be significant variables explaining C-OCB, while Tenure (β=.16, p < .05) and PO fit (β=.51, p < .001) were once again found to be statistically significant.
To conclude the main effect of PO fit on C-OCB, model 11 was developed. In this model, all cultural types and PO fit were entered as independent variables, and C-OCB was entered as the dependent variable. This model explained 31% extra variance on top of the control variables (R2 change= .31; F(5,129)=9.7; p < .01). PO fit was the only significant predictor of C-OCB in this model (β=.37, p < .01).
25 Table 2. Summary of Hierarchical Analyses for Variables Predicting C-OCB (N=135)
Variables Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7 Model 8 Model 9 model 10 Model 11
Tenure .18* .19* .18* .25** .17* .19* .17* .17* .19* .16* .18 Contract Hours .12 .08 .12 .14 .12 .12 .13 .10 .14 .14 .13 Clan .41** .33** .12 .12 Internal Process -.01 -.03 .03 .08 Ad Hoc .35** .15 .10 .07 Market -.05 -.12 -.09 -.13 PO fit .33** .46** .44** .51** .37** Clan X PO fit -.156
Internal Process X PO fit -.15
Ad Hoc X PO fit .04 Market X PO fit .08 R2 .05 .22 .05 .17 .05 .24 .31 .30 .29 .29 .56 F 3.5* 12.6** 3.58 9.1** 2.5 6.9** 11.9** 11.3** 10.6** 11.1** 8.29** R2 change .17 .00 .12 .00 .19 .02 .02 .00 .00 .31 F change 29.1** .01 19.26** .38 8.2** 3.3 3.6 .26 1.1 9.7**
Note: β coefficients are stated
** Correlation is significant at the 0.01 level (2-tailed) * Correlation is significant at the 0.05 level (2-tailed
26 5 Discussion
The aim of this chapter is to provide an overview of the research and discuss the theoretical and practical implications of the findings. Furthermore, limitations of the analysis are evaluated and suggestions for future studies are made. Lastly, a brief conclusion is given.
5.1 Summary
The purpose of this study is to explore the relationship between organizational culture and C-OCB. C-OCB is becoming increasingly important in organizations and therefore understanding its antecedents may enable us to gain a better insight in how to promote change within an organization. If it is evident which organizational cultures lead to change-oriented behaviors, organizations can focus on emphasizing such cultures. This will make organizations more flexible and change-oriented, which is crucial for thriving in the current business environment. In this study, it was hypothesized that three cultural types relate to OCB. First, it was expected in Hypothesis 1a that the clan culture is positively linked to C-OCB. The results indeed show a positive relationship between the clan culture and C-C-OCB. Hypothesis 1a is thus confirmed. This means that employees who work in a clan culture are likely to elicit change-oriented citizenship behaviors. Secondly, Hypothesis 1b predicted that a negative relationship exists between the internal process culture and C-OCB, but the analysis could demonstrate no significant relationship between the two. Therefore, Hypothesis 1b is not confirmed. Third, Hypothesis 1c claimed that the ad hoc culture is positively related to C-OCB. The research initially pointed towards a significant effect, suggesting that ad hoc cultures encourage C-OCB. However, the significance of this effect disappeared in the multi-stage hierarchical regression when the factor representing clan culture was added to the model. Therefore, the findings do not support Hypothesis 1c. Due to contradictory previous research, no directional hypothesis could be construed regarding the relationship between the market culture and C-OCB. Therefore it was expected no significant relationship would be found. This assumption was supported by the statistical analysis, which showed no significant relationship between the two.
In addition to examining the impact of the main independent variables on C-OCB, this study also investigated whether PO fit is a moderator in the relationship between
organizational culture and C-OCB. PO fit was expected to act in such a way that a high PO fit would strengthen the relationship between the cultural type and C-OCB and a low PO fit
27 would weaken this relationship. Hypotheses 2a, 2b, and 2c assert that PO fit moderates the relationship between organizational culture and C-OCB. The results of the moderation analyses show no support for these claims. Thus it is found that PO fit does not moderate the relationship between organizational culture and C-OCB.
5.2 Theoretical implications of Direct effects
The results indicate that employees who work in a clan culture are more likely to show change-oriented citizenship behaviors. The clan culture distinguishes itself from others by its people-orientation. It has a strong focus on support, trust, and participation. The norm
of reciprocity (Gouldner, 1960) predicts that people are likely to repay, in kind, that what
other people have provided for them. In the clan culture, employees receive much support from other employees and the organization. Based on the norm of reciprocity, one might expect that employees, therefore, repay this support by displaying change-oriented
citizenship behavior. C-OCB may then be interpreted as an obligation an employee owes to the organization. Even though these different characteristics of the clan culture have
previously been linked to C-OCB (Chiaburu et al., 2013; Choi, 2007), this is the first study to find a positive relationship between a clan culture and C-OCB. The confirmation of this hypothesis raises new questions for research. How does management encourage and facilitate a clan culture? Moreover, is it a goal worth pursuing taking into account the needed resources to facilitate such a culture? If C-OCB can be attributed to certain elements of a clan culture, it might only be necessary to implement these specific elements in their organizations to
encourage C-OCB. Future research could investigate the relationship between perceived organizational support, participation and trust and C-OCB. This could help identify the most important elements that account for the positive relationship between the clan culture and C-OCB.
The internal process culture is concerned with routinization and formalization, both of which have been found to be negatively related to C-OCB (Marinova et al., 2015).
Employees are expected to follow strict rules and procedures in the internal process culture. Clearly defined role descriptions are provided, which limits the opportunity to engage in C-OCB in this type of culture. It was therefore expected in this study that the internal process culture would be negatively related to C-OCB. However, the relevant coefficient was not significant. This challenges previous studies since there was strong evidence suggesting a negative correlation. For example, Marinova et al. (2015) found a negative relationship
28 between both formalization and centralization and C-OCB. These are both main features of the internal process culture. Since C-OCB is self-initiated proactive behavior, it was expected there would be no room for C-OCB in a culture that has strict rules and clear role
descriptions.
An explanation for not finding a negative relationship between the two might be that employees often view OCB as an aspect of their in-role performance. (NahumShani &
Somech, 2011). If one drops the assumption that C-OCB is per definition extra-role behavior, it is possible that clear role descriptions promote C-OCB. For example, if company X has clearly defined role descriptions that state employees should be on the lookout for
improvements constantly, this might promote C-OCB since employees know what is
expected of them. Future research could explore the question whether or not change-oriented behaviors can be promoted by adding such behaviors to the role descriptions and regulations. If so, organizations can promote C-OCB by simply stating the wanted behavior.
Another possible explanation is that other variables are in play next to the internal process culture that positively relate to C-OCB. For example, company Y can have a powerful internal process culture since it is a manufacturing company whereby safety, optimal processes, and cost control are crucial. On top of this, employees at company Y are taken outstanding care of. Therefore, the employees perceive much organizational support. According to the social exchange theory, when employees feel like the organization cares about their well-being they feel obligated to engage in extra-role behaviors such as C-OCB (Cropanzano & Mitchell, 2005). Therefore even organizational cultures that are very much internal process focused may be abundant with C-OCB. Perhaps the characteristics of an organizational culture that negatively relate to C-OCB do not influence employees as long as they perceive sufficient organizational support. Future research could look into possible variables that affect the relationship between an internal process culture and C-OCB, such as perceived organizational support.
The ad hoc culture is characterized by autonomy, leadership, strong vision and personal initiative. These characteristics are strongly linked to extra-role behaviors
(Binnewies & Gromer, 2012; Marinova et al., 2015; Parker et al., 2006). The results initially showed a significant positive relationship between the ad hoc culture and C-OCB, but it lost its statistical significance once the clan culture was added to the model. This might be explained by the fact that there is a significant correlation between Clan and Ad Hoc (r (135)=.70, p <.05), as demonstrated in Table 1. In other words, the initial significance obtained for the variable Ad Hoc could be attributed to its high level of correlation with the
29 variable Clan. Both the clan and the ad hoc cultures are part of the flexible dimension of the Competing Values Framework. Although correlations between the cultural types located within the same framework might be expected, a correlation of .70 between Clan and Ad Hoc is somewhat unusual because the former has an internal focus whereby the latter has an outward one.
In contrast to the findings of previous research studies, this research could not confirm the presumed positive relationship between an ad hoc culture and C-OCB. This has important implications since it is widely assumed that cultures which promote autonomy and strong vision stimulate C-OCB (Binnewies & Gromer, 2012; Marinova et al., 2015; Parker et al., 2006; Zaccaro & Banks, 2004). However, according to this study, such a relationship does not exist. Perhaps other factors, unrelated to culture have a stronger influence on OCB, and therefore diminish the positive relationship between the ad hoc culture and C-OCB.
There is much sound theoretical evidence suggesting a positive link between the ad hoc culture and C-OCB. Therefore it might be possible that methodological issues caused this non-significant result. A possible explanation relates to the survey design as the survey statements for both cultural types were right beneath each other in the questionnaire. Some participants, in fact, noted that they had difficulties in differentiating the questions of the CVF, complaining that “the statements all looked very similar”. In fact, the average of the means of the different cultural types is very high (m=4.42), with the means varying between 4.16 and 4.89.
The ad hoc culture’s focus on autonomy provided the most important rationale for the hypothesis that predicted a positive relationship with C-OCB, since the literature provides strong evidence that autonomy is an important determinant of C-OCB (Frese et al., 1997 Van Dyne & Lepine, 1998). The survey had four statements with 13 characteristics to assess the ad hoc culture, but only one related to autonomy by utilizing the term ‘freedom’. The survey also did not measure the presence of vision, another characteristic that is also believed to be strongly related to C-OCB. Thus, what explains the failure of the ad hoc culture hypothesis might perhaps be the fact that the research survey did not have all-inclusive measurements to properly assess the ad hoc culture when entered together with the other culture items.
In summary, out of the three hypotheses formulated regarding the relationship between organizational culture and C-OCB, only one, namely clan culture, was found to be significantly related to C-OCB. PO fit was analyzed as a possible moderator in these relationships. The results of the moderation analyses are discussed in the next section.
30 5.3 Theoretical implications of Moderation effects
Organizational culture can be seen as a value framework that has distinct boundaries defining appropriate behaviors. Therefore, it was expected that employees with a high PO fit could better understand and embrace the signals about appropriate behaviors transmitted by the organization. Subsequently, it was postulated that PO fit moderates the relationship between organizational culture and C-OCB in such a way that employees with a high PO fit make the relationship stronger while employees with a low PO fit make it weaker. However, this research could produce no evidence supporting any of the PO fit hypotheses. It should be noted that the PO fit variable achieved significance in the regression model when its impact on C-OCB is considered along with all the cultural types.
PO fit was found to be strongly correlated with both the clan culture (r(135)=.67, p < .001), and the ad hoc culture (r(135)=.59, p<.001). PO fit was also strongly correlated with C-OCB (r(135)=.48, p <.001). These facts suggest that PO fit might be acting more as a mediator than a moderator in the relationship between organizational culture and C-OCB. This could imply that employees who work in more flexible organizational cultures might be more likely to have a higher PO fit. The argumentation for this alternative model is that these cultures, when compared to the internal process and the market culture, have a stronger focus on a positive relationship with their employees. In the clan culture, this entails support and trust while in the ad hoc it entails autonomy and freedom. Employees who work in a clan or ad hoc culture show a higher degree of PO fit probably because their organizations seem to care about them more.
To test this alternative model, a mediation test was conducted with a custom dialog called PROCESS in SPSS. Clan was entered as the independent variable, PO fit as the mediator and C-OCB as the dependent variable. All other cultural types were added as control variables. The results showed a positive correlation between clan and PO fit (β=.55,
p < .01). Furthermore, a positive relationship was found between PO fit and C-OCB (β=.33, p
< .01). The main effect of clan on C-OCB was not significant (β=.16, p > .05). This means
PO fit fully mediates the positive relationship between clan and C-OCB. Thus, in this
relationship PO fit acts as a mediator. These results support the alternative model in the previous paragraph. This implies that PO fit does not alter the relationship between organizational culture and C-OCB, but rather explains the relationship. According to this alternative model, the positive relationship between a clan culture and C-OCB is due to a higher PO fit in this culture.
31 Another alternative explanation is that PO fit is simply a much stronger predictor of C-OCB than is organizational culture. An organizational culture is construed of many different characteristics. Some of these features might positively influence C-OCB, while others within this same culture might not. Furthermore, individual responses of the
employees regarding a specific characteristic, such as autonomy, might affect the relationship between the characteristic and C-OCB. For example, one employee might thrive with much autonomy and engage heavily in C-OCB, while the effect of autonomy on another employee might be completely different. These arguments might show that the relationship between organizational culture is rather indirect and furthermore depends upon individual responses of employees. PO fit, on the other hand, has been found to have a strong positive relationship with OCB (de Lara, 2007; Farzaneh et al.,2014; Wei, 2013; Vilela et al., 2008). Perhaps PO fit is a stronger predictor since it is a direct measure regarding the employee’s beliefs, and therefore directly influences employee behavior. The theoretical implication of this would be that the perceived value congruence between an employee and their organization is more important in predicting C-OCB, than the type of prevalent culture they work in.
5.4 Limitations and future research
The results of this research should be interpreted with some caution due to its certain limitations. First, it used a cross-sectional research design, which does not allow to reach sound conclusions regarding the causality of the variables. The directions of the relationships were therefore inferred from the theoretical arguments and the literature in the study. Future research focusing on longitudinal effects may be advised to strengthen the findings of this research and provide a more empirical reasoning regarding the direction of the relationships.
Secondly, the data for this study was collected solely using self-reported measures. This may lead to certain estimation biases such as the common method bias or the social desirability effect (Podsakoff & Organ, 1986). In the case of self-measuring OCB, there is a positive bias, and people tend to rate their OCB higher than others (Organ & Ryan,
1995). Future research might include additional measures based on third-party ratings like supervisor measurements and ratings.
Third, there are some limitations regarding the sample. The sample used for this study is not an accurate representation of the population of employees in the workforce of The Netherlands. Due to inherent issues of convenience sampling, the participants in the sample turned out to be highly homogenous, exhibiting little variation in their educational and
