Knowledge Inflows and Managers’ Exploration and Exploitation
Activities: The Moderating Role of Managers’ Need for Cognitive
Closure
Master’s Thesis
Student:
Eliane Rikkers/
Student’s № 6070264/University of Amsterdam, Faculty of Economics and Business Business studies: Strategy
Supervisor: Dhr. Bernardo Silveira Barbossa Correia Lima
ABSTRACT
This study was motivated by a wish to develop a greater understanding of the exploration and exploitation activities at a manager level of analysis, since existing research on exploration and exploitation activities has focused mainly at the organizational or business-unit level of analysis. The primary goal of this thesis was to examine the influence of intra-organizational knowledge flows on managers’ exploration and exploitation activities, while taking into account managers’ motivation to absorb and use the knowledge measured by the level of need for cognitive closure (NFCC). An online survey was sent to 277 middle level managers employed by ABN AMRO, one of the largest banks in the Netherlands, and was filled out by 100 respondents. Data analyses revealed that horizontal knowledge inflows are an important driver for the exploration activities managers engage in, while it did not find any significant results for bottom-up and top-down knowledge inflows on exploration and exploitation. Moreover, it seems that while NFCC does not reveal a moderating impact on the relationships between intra-organizational knowledge and managers’ exploration and exploitation activities, it does provide us with a few unexpected, though important results. Managers with high levels of NFCC tend to engage more in exploitation activities compared to managers with a low level of NFCC. Furthermore, it seems that NFCC is negatively related to bottom-up and horizontal knowledge inflows. The results offer high-yield contributions to both theory and practice, as well as avenues for future research. Overall, it brings us a step closer in understanding exploration and exploitation activities on a manager level of analysis.
Keywords: exploration and exploitation; knowledge inflows; need for cognitive closure; manager level; motivation
TABLE OF CONTENTS
ABSTRACT
I Introduction 1
II Theoretical background 6
1. Managers’ exploration and exploitation activities 6
2. Knowledge inflows 7
3. Managers’ need for cognitive closure 10
III Hypotheses development 13
1. Bottom-up knowledge inflows and managers’ exploration and exploitation activities 13 2. Horizontal knowledge inflows and managers’ exploration and exploitation activities 15 3. Top-down knowledge inflows and managers’ exploration and exploitation activities 16
4. Moderating effects of need for cognitive closure 18
4.1 Bottom-up and horizontal knowledge inflows and managers’ exploration activities: The
moderating effect of need for cognitive closure 18
4.2 Top-down knowledge inflows and managers’ exploitation activities: The moderating
effect of need for cognitive closure 20
5. Conceptual model 22
IV Methodology 23
1. Research design 23
2. Sample and data collection 24
3. Measurements 25 3.1 Dependent variables 26 3.2 Independent variables 28 3.3 Moderator 30 3.4 Control variables 31 V Results 34 1. Missing values 34 2. Descriptive statistics 34 3. Correlation analysis 37 4. Regression analysis 40 4.1 Assumptions 40 4.2 Main effects 41 4.3 Moderation effects 45 4.4 Unexpected effects 48 VI Discussion 50 1. Research findings 50
1.1 Main effects: knowledge inflows and exploration and exploitation 50
1.2 Moderation impact: need for cognitive closure 52
1.3 Control variables 53
2. Implications for theory 54
3. Implications for practice 56
4. Limitations and avenues for future research 57
VII Conclusion 60
References 62
Appendices
TABLES
Table 1. Items exploration activities 26
Table 2. Items exploitation activities 27
Table 3. Items knowledge inflows 29
Table 4. Items need for cognitive closure 31
Table 5. Descriptive statistics: continuous variables 35
Table 6. Descriptive statistics: categorical variables 36
Table 7. Means, standard deviations, correlations 39
Table 8. Hierarchical multiple regression analysis. Bottom-up, horizontal and top-down knowledge
inflows with dependent variable: Exploration 43
Table 9. Hierarchical multiple regression analysis. Bottom-up, horizontal and top-down knowledge
inflows with dependent variable: Exploitation 45
Table 10. Moderation: need for cognitive closure and bottom-up knowledge inflows 46 Table 11. Moderation: need for cognitive closure and horizontal knowledge inflows 47 Table 12. Moderation: need for cognitive closure and top-down knowledge inflows 48 Table 13. Items and factor analysis of managers’ exploration and exploitation activities 71 Table 14. Items and factor analysis of bottom-up, horizontal and top-down knowledge flows 72
I Introduction
In the dynamic world we live in today, it remains a big challenge for firms to adapt to the changing environment and keep up and stay in the market. Organizations need to explore opportunities, but also exploit that which is already known (March, 1991) in order to be able to survive. Exploration and exploitation activities are both essential to an organization, but quite the opposite from each other. Exploration activities involve “searching for, discovering, creating and experimenting with new opportunities” (March, 1991, in Mom, Van Den Bosch & Volberda, 2007, p. 910) while exploitation activities are about “selecting, implementing, improving and refining existing certainties” (March, 1991, in Mom, Van Den Bosch & Volberda, 2007, p. 910).
The existing research on exploration- and exploitation activities mainly focuses on the organizational level (Benner & Tushman, 2002; March, 1991; Harreld, O’Reilly & Tushman, 2007; Jansen, Van Den Bosch, & Volberda, 2006; Lavie, Stettner & Tushman, 2010). There is, however, a lack of insight into the exploration and exploitation activities at a management level (Gibson & Birkenshaw, 2004; Mom, Van Den Bosch & Volberda, 2007, 2009). This is quite peculiar as the exploration and exploitation activities managers engage in are often partly an indication of the organization or unit level exploration and exploitation activities (Burgelman 1983b, 1991; Floyd & Lane, 2000, in Mom et al., 2007). This thesis contributes to existing literature by taking a close look at the exploration and exploitation activities at a manager level of analysis.
Research indicates that knowledge flows could be associated with exploration and exploitation activities. It seems that knowledge transfer, by combining existing and new knowledge allows firms to generate new innovative ideas – exploration – (Powell et al., 1996), while it also gives firms a better sense of emerging needs to adapt current products and services – exploitation – (Day, 1994). Individuals form ideas in their mind, but the interaction between people play an essential role in developing these ideas (Nonaka, 1994). It is therefore fundamental that knowledge is being shared inside an organization (Kogut & Zander, 1992). However, knowledge that comes from lower level employees – bottom-up knowledge inflows – differs fundamentally from knowledge that is
acquired from higher level employees – top-down knowledge inflows – or from employees at the same hierarchal level – horizontal knowledge inflows – and thus will have differing impacts on the exploration and exploitation activities that managers engage in.
Although previous research has examined the relationships between intra-organizational knowledge inflows and managers’ exploration and exploitation activities, no research has ever considered the motivation of individuals to listen to and use these particular knowledge inflows for their exploration or exploitation activities. Motivation is especially likely to have an impact on these relationships, since previous research suggested that individuals differ in their degree of knowledge acquisition and knowledge use, because of differences in motivation. It seems that the motivation to absorb information, creates differences in the way information is processed, which in turn could lead to differences in acting upon this information (Minbajeva et al., 2012). A notion that looks at an individual’s motivation with respect to absorbing knowledge, knowledge processing and judgment is the ‘need for cognitive closure' (Webster & Kruglanski, 1994).
Need for cognitive closure (NFCC) is a concept defined as “a desire for a definite, firm answer to a question, in contrast to tolerance of uncertainty, confusion, or ambiguity with respect to that question” (Kruglanski 1989, p. 337). Individuals with a high NFCC want to find a solution to a problem as quickly as possible which in turn could lead to biases in their information processing (Kruglanksi & Webster, 1996). These individuals generally are close-minded, feel uncomfortable dealing with ambiguity and most importantly make use of a shallow knowledge search (Bukowski, Von Hecker & Kossowska, 2013; Webster & Kruglanki, 1994). On the contrary, individuals with a low NFCC are more willing to acquire new knowledge, to look at all the different knowledge that is presented to them and thus are inclined to make use of a broader knowledge search (Van Hiel & Merfielde, 2003). This study will go beyond existing literature on exploration and exploitation activities at the individual level, by examining intra-organizational knowledge inflows and managers’ exploration and exploitation activities, while taking into account managers’ NFCC.
A major reason to believe that NFCC could have an impact on the relationships between intra-organizational knowledge flows and managers’ exploration and exploitation activities, is the notion that extant literature revealed conflicting results regarding these relationships. The variance in motivation for listening to and using different knowledge inflows might explain the different research results (Mom et al., 2007; Wei et al., 2011). Research by Wei et al. (2011) was conducted in China, while research by Mom et al. (2007) was conducted in Europa. Previous research found that Asian individuals have a significantly higher NFCC than European individuals (Kossowoska & Van Hiel, 2002). It is thus possible that Wei et al. (2011) conducted research among managers with a higher NFCC, resulting in different effects, compared to the lower level of NFCC from managers surveyed by Mom et al. (2007).
To understand the impact NFCC could have, there first ought to be an understanding of how knowledge flows could influence different activities and what the different studies revealed. In order for managers to engage in exploration activities, they need knowledge that will broaden their existing knowledge base, whereas for exploitation activities they need knowledge that will deepen their existing knowledge base (Levinthal & March, 1993). Broadening a knowledge base offers managers new, ambiguous and varied knowledge to recombine with existing knowledge, so new opportunities can be explored. Deepening a knowledge base adds knowledge that is closely related to the pre-existing knowledge base, so knowledge can be reused to refine existing certainties – exploitation (Katila & Ahuja, 2002).
Mom et al. (2007) found that top-down knowledge inflows had a positive impact on exploitation. It was argued that this type of knowledge flow generates more specialized and understandable knowledge that deepens the manager’s existing knowledge base. In addition, bottom-up and horizontal knowledge inflows were positively related to managers’ exploration activities, since these knowledge flows offer new, ambiguous and varied knowledge that will broaden a manager’s existing knowledge base. On the contrary, Wei et al. (2011) found a positive association between bottom-up knowledge inflows and manager’s exploitation activities and an
inverted u-shaped effect on exploration activities. NFCC could have a moderating impact on these relations and thereby reconcile the contradictory results from previous research.
It is assumed that research by Wei et al. (2011) was conducted among individuals with a higher level of NFCC compared to research by Mom et al. (2009) and thus their results might have deviated. Wei et al. (2011) found that bottom-up knowledge inflows started with a positive effect on exploration activities, but this effect declined along its increasing. Individuals with a high NFCC dislike uncertainty, confusion and ambiguity, which are exactly the characteristics of the bottom-up knowledge flows that broaden a knowledge base for exploration (Katila & Ahuja, 2002). Wei et al. (2011) their finding could be explained by the notion that a high NFCC could lead to less motivation to deal with these broad and vague knowledge inflows and thus are used less for exploration activities. Having a low level of NFCC is associated with an increased willingness to acquire new and ambiguous knowledge, which could clarify why Mom et al.’s (2007) managers showed a positive relationship between bottom-up knowledge flows and their exploration activities.
Furthermore, Wei et al. (2011) found that bottom-up knowledge inflows are positively associated with exploitation, while Mom et al. (2009) found no effect for this relation. A higher NFCC is assumed to be associated more with refining existing activities – exploitation – than experimenting with new opportunities – exploration, since high NFCC individuals have an aversion to the uncertainty and ambiguity that exploration activities often create (Kruglanski & Webster, 1996). It is thus argued that the positive relationship between bottom-up knowledge inflows and exploitation activities (Wei et al. 2011) might not be the result of bottom-up knowledge inflows, but instead is the effect of NFCC that is more associated with refining day-to-day activities than exploring new opportunities.
This paper seeks to address the relationships between knowledge inflows and managers’ exploration and exploitation activities, while taking into account a managers’ level of NFCC. NFCC is expected to have different moderating impacts on these relationships, since the knowledge flows differ with regard to their characteristics. This will be further elaborated on in following chapters.
There needs to be a deeper understanding of what causes managers to explore and/or exploit more. Hence, the research questions that this thesis aims to answer are:
What are the relationships between (bottom-up, horizontal and top-down) knowledge inflows and managers’ exploration and exploitation activities?
What is the moderating effect of managers’ need for cognitive closure on these relationships?
By empirically examining these associations, this study makes several contributions to both theory and practice. First, it will create a deeper understanding of managers’ exploration and exploitation activities as only limited research has been done at the manager level of analysis. Second, this study will go beyond existing literature on exploration and exploitation at an individual level by examining the moderating impact of the motivation to listen to and use particular knowledge flows for exploration or exploitation activities. This moderator variable can resolve contradictory findings and gain new insights into the process of knowledge flows. Third, it can uncover practical implications for the hiring of particular types of managers, stimulation of certain knowledge inflows to get managers to explore or exploit more, knowledge development and ultimately for a firm’s exploration and exploitation activities. Finally, identifying unexplored questions and limitations of the existing research will highlight promising areas for future research.
This thesis will be structured as followed. First the literature review will discuss the main insights on the important concepts. Next, the theoretical framework will point out the subsequent hypotheses. Furthermore, the method will be outlined and the main findings will be presented in the results section. Fourth and final, this paper will be concluded by discussing the contributions to the existing literature, as well as implications for management practice and avenues for future research.
II Theoretical background
The following paragraphs discuss the main insights of the existing literature on the most important concepts. First the exploration and exploitation activities of managers will be introduced. Second, the main literature on bottom-up, horizontal and top-down knowledge inflows will be presented. Third, the moderating role of a manager’s ‘need for cognitive closure’ will be discussed.
1. Managers’ exploration and exploitation activities
Exploration and exploitation activities differ fundamentally from each other. They are nevertheless both needed for a firm to survive in the constantly changing environment. Organizations need to balance the differing demands for efficiency in the short term and effectiveness in the long term (March, 1991). There must be a thorough understanding of what causes managers to exploit or explore more, since the exploration and exploitation activities managers engage in is often an indication of the exploration and exploitation activities on a business-unit or organizational level of analysis.
March (1991, p. 71) considers exploration activities as “things captured by terms such as search, variation, risk taking, experimentation, play, flexibility, discovery and innovation” while exploitation activities are “things as refinement, choice, production, efficiency, selection, implementation, and execution”. In addition, exploration activities are related to broadening the existing knowledge base whereas managers’ exploitation activities are related to deepening the existing knowledge base (Katila & Ahuja, 2002; Levinthal & March, 1993; Mom, Van Den Bosch & Volberda, 2007). In other words, exploration concerns “a pursuit of new knowledge” whereas exploitation concerns “the use and development of things already known” (Levinthal & March, 1993, p. 105). As those activities are so different from each other, the processes, routines and skills required for exploration and exploitation also differ fundamentally (O’Reilly & Tushman, 2004).
This thesis focusses on the exploration and exploitation activities middle level managers engage in. Middle level managers are defined as the employees who are at least two levels below
the CEO and at least one level above first-line supervisors (Huy, 2002). Middle level managers are regularly seen as employees that need to maintain continuity and are often busy with implementing and executing existing activities (Huy, 2002). According to most scholars, they therefore mainly engage in exploitation activities. Newest research, however, indicates that managers are not only worthy for exploitation activities, but can also form a strategic asset and become involved in exploration activities (Floyd & Wooldridge, 1992, 1994, 1997, in Balogun, 2003). Middle level managers can make valuable contributions to new strategies in an organization because they often have worthy entrepreneurial ideas and, more important, because they are regularly better than top management in leveraging their informal networks (Huy, 2001). Middle level managers have an advantageous organizational position since they have access to knowledge from multiple levels of the organization.
2. Knowledge inflows
Knowledge is seen as a valuable resource. Many different studies in the field of strategy, such as the resource-based view and the knowledge-based view, see knowledge as a source of sustained competitive advantage as it is a unique, path dependent, causally ambiguous, intangible asset that is hard to imitate or substitute (Diederikx et al., 1989; Grant, 1996; Peteraf, 1993; Teece et al., 1997). This illustrates the importance of employees because without employees knowledge cannot be developed within an organization (Lahti & Beyerlein, 2000). Nevertheless, employees need to share their knowledge to drive new knowledge creation. Knowledge sharing is the process of making knowledge available to other individuals (Ipe, 2003) as it transfers know-how within the firm, from where it is known to where it is not (Hargadon, 1998).
Knowledge inflows of managers is defined as “the ‘aggregate volume’ (Schulz, 2001, p. 662) of tacit and explicit knowledge pertaining to several domains such as technology, products, processes, strategies and markets, which a manager receives or gathers per unit of time, from other persons and units within the organization” (Mom, Van Den Bosch & Volberda, 2007, p. 913). Transferring knowledge inside an organization is seen as more effective and efficient due to better recognition
and less disclosure problems than knowledge that comes from outside the organization (Gupta & Govindaranjan, 1994). According to previous research, intra-organizational knowledge flows offer competitive benefits (Gupta & Govindarajan, 2000; Schulz, 2001) as they are generally associated with firm performance and innovation (Van Wijk, Janssen & Lyles, 2008). However, research has focused mainly on a firm level of analysis and thereby neglected the essential role of individuals in knowledge processes (Argote & Ingram, 2000).
This thesis focusses on intra-organizational knowledge flows acquired by middle level managers, which could be distinguished by bottom-up, top-down, and horizontal knowledge inflows. Managers’ top-down knowledge inflows consist of the knowledge that is received from their supervisor or employees at an even higher hierarchical level than their supervisor. Managers’ bottom-up knowledge flows comprise the knowledge that comes from employees near the front line or front-line supervisors. Lastly, managers’ horizontal knowledge inflows come from managers at the same hierarchical level, either within the same organizational unit or from other units (Mom et al., 2009). These three types of knowledge inflows are profoundly different from each other as they have different attributes. The attributes of knowledge flows are considered as an important antecedent to knowledge transfer (Zander & Kogat, 1995, in van Wijk, Janssen & Lyles, 2008) and thus might generate different effects.
Concerning bottom-up knowledge inflows, the most important notion is that frontline employees interact with costumers (Ye, Marinova & Singh, 2012). Frontline employees frequently need to improvise solutions by working with knowledge involving changing customer needs and market conditions, unforeseen problems and new technological developments (Burgelman, 1983b; Mom et al., 2007; Van De Ven, 1980; Ye, Marinova & Singh, 2012). These knowledge flows generally do not come about in a standardized or formalized way since they are often random, impulsive and unplanned, and usually call for qualitative instead of quantitative changes (Burgelman, 1983b; Sanchez & Heene, 1996, in Mom et al., 2007). Hence, the knowledge that comes from lower level employees is often new and not closely related to managers’ existing knowledge base. In addition,
research indicates that employees at the operational levels of a firm often come up with autonomous strategic initiatives that could be the motor of corporate entrepreneurship (Burgelman 1983b). If this knowledge is collected and captured, it could be a great source of competitive advantage (Grant 1996; Nonaka, 1994, in Ye, Marinova & Singh, 2012).
Horizontal knowledge flows seem to be somewhat similar to bottom-up knowledge inflows since it also generates knowledge that is not closely related to managers’ existing knowledge base. These knowledge flows are generally received through other parts of the organization, in which the manager is less familiar with and it tends to be about unknown processes and activities. The knowledge flows are therefore expected to be quite broad, ambiguous, tacit and complex (Daft & Lengel, 1986; Egelhoff, 1991). Tacit knowledge is knowledge that only exists in the head of individuals and cannot be codified (Polanyi, 1962, in Minbaeva, 2010). This type of knowledge is more complex to acquire but is also seen as a competitive advantage, due to the difficulty of imitating it. Another important factor of horizontal knowledge flows is that it is mainly acquired though cross-functional personal interactions (Tsai, 2001). This brings in the social aspects of exchanging knowledge (Williamson, 1985).
Top-down knowledge flows are more distinctive from bottom-up and horizontal knowledge flows. These knowledge inflows are restricted to the vertical chain in the area of expertise and organizational unit the manager is situated in (Gupta & Govindarajan, 1994; Hedlund, 1994, in Mom et al., 2007). In other words, managers will mainly receive knowledge from above about the area they are specialized in and this knowledge will be more directed towards known activities. Hence, these knowledge flows are less new and diverse and more related and directed towards managers’ existing knowledge. As a consequence, the top-down knowledge flows usually are expected to be understandable and clear and do not come about in an unplanned or unstandardized way(Egelhoff, 1999; Winter & Szulanki, 2001). To conclude, characteristics of top-down knowledge inflows are profoundly different from the characteristics of bottom-up- and horizontal knowledge inflows. It is therefore important to have a deeper understanding of the different impacts these knowledge flows
could have on the different activities managers engage in.
Managers are often in the middle of an organization and thus have important connections and unique access to top management, peer managers in different units and employees near the front line (Nonaka, 1994). They are assumed to be the link between the strategic apex and the operating core (Thompson, 1967, in Balogun, 2003) and are able to connect these two levels through negotiation and interpretation (Floyd & Wooldrigde, 1997). Hence, managers have an advantaged organizational position from where they can access a large spectrum of diverse knowledge to achieve knowledge creation (Tipmann, Mangematin & Scott, 2013). Their privileged position within the organization’s structure permits them to retain social relationships with a variety of employees who are all potentially worthy for accessing this diverse and new knowledge (Mors, 2010; Pappas & Wooldridge, 2007, in Tippmann, Mangematin & Scott, 2013). Although managers have a superior potential to access knowledge, it does not mean they will also realize this potential. Having a ‘knowledge access advantage’ is different from having a ‘knowledge use advantage’ defined by Tipmann et al. (2013, p. 1873) as: “the taking up of search opportunities in practice, and the actual use of knowledge made available or supplied by individuals and/or the wider organization”. Not having the motivation to use particular knowledge and thereby processing knowledge differently could be a reason why managers receive certain knowledge inflows but not follow up on the potential of this knowledge. For that reason, a moderating role of managers NFCC will be examined.
3. Managers’ need for cognitive closure
Individuals with a high level of NFCC have a habit of making a shallow analysis of incoming knowledge and the motivation to only accept information that is consistent with their existing beliefs (Bukowski, Von Hecker & Kossowska, 2013). This concept implies that there is a difference in individuals’ motivation to engage in complex thinking processes (Neuberg, Judige & West, 1997). Individuals with a high NFCC prefer order and predictability, are decisive, close-minded and feel uncomfortable when dealing with ambiguity (Webster & Kruglanski, 1994). The opposite of individuals with a high NFCC are individuals with a high need to avoid closure. These individuals have
the desire to defer judgmental commitment and make use of a broad information search when looking for solutions (Van Hiel & Merfielde, 2003; Webster & Kruglanki, 1994). The longing for cognitive closure varies along a continuum where a strong need for cognitive closure exists at one end and a strong need to avoid closure exists at the other end (Kruglanski & Webster, 1996).
There are two underlying tendencies in having a high NFCC: the urgency tendency and the permanence tendency (Kruglanksi & Webster, 1996). The urgency tendency refers to the tendency to ‘seize’ on closure quickly. Individuals with this tendency want the closure to happen as soon as possible and feel there will be drawbacks when postponing it. The permanence tendency refers to the tendency to maintain the closure for as long as possible. These tendencies make sure that individuals seize on information that appears early and instantly freeze on it so no additional data is needed. These seizing and freezing trends have detrimental effects on what information is processed and how the information is processed (Kruglanski & Webster, 1996). Dual process theories explain why people reason and process information differently.
Dual process theories make a distinction between heuristic processing and analytical processing. Heuristic processing is considered as being quick, automatized and based on existing beliefs and knowledge. Analytic processing, on the other hand, is described as being less fast, more effortful and demands integration of receiving pieces of information (Evans, 2008, in Bukowski, Von Hecker & Kossowska, 2013). It is assumed that people with high NFCC rely mostly on heuristic processing whereas people with low levels of NFCC rely more on logical, analytical reasoning (Bukowski, Von Hecker & Kossowska, 2013; Kruglanksi, 1989).
Multiple studies show that NFCC invites biases in information processing. Bukowski, Von Hecker and Kossowska (2013) found that high NFCC individuals trust mostly on the first information presented when forming an impression and try to draw a quick and simple conclusion. They rely broadly on available knowledge structures and heuristic processing in contrast with lower NFCC individuals who rely more on logic and analytical reasoning (Bukowski et al., 2013; Kruglanski, 1989). Furthermore, Van Hiel and Merfielde (2003) found that high NFCC individuals use a narrow and
shallow information search when they are trying to find a solution to an ambiguous situation. Also, high NFCC individuals accept information that challenges their view significantly less than information that supports their view (Hart et al., 2012). The consequence is that high NFCC individuals process less information before they feel committed to an opinion and bring forth fewer opposing hypotheses to account for the available data (Kruglanksi & Webster, 1996). In addition, they feel more confident about their judgment because less competing hypotheses lead to fewer alternatives to contradict their judgments (Kelly et al., 1971). In sum, the NFCC creates biases in selecting the most relevant information for a decision, assessing and judging the input information and lastly weighting this information to make the right decision. Hence, managers with a high NFCC will probably process particular knowledge inflows differently from managers with a low NFCC.
III Hypotheses development
The following paragraphs will discuss the relationships between the main concepts and the subsequent hypotheses. First, the relationship between bottom up knowledge inflows and managers’ exploration and exploitation activities will be discussed. Second, the relationship between horizontal knowledge inflows and managers’ exploration and exploitation activities will be conferred. Furthermore, the relationship between top-down knowledge inflows and managers’ exploration and exploitation activities will be presented. Fourth and final, the moderating effects of managers’ NFCC on these relationships will be explored.
1. Bottom-up knowledge inflows and managers’ exploration and exploitation activities
Bottom-up knowledge inflows refer to knowledge that comes from employees at a lower hierarchical level than the manager. Knowledge that is resided in employees who are near the frontline is different from the knowledge that is resided in employees at the same or higher hierarchical levels.
In order for managers to engage in exploration activities, new knowledge has to be added to broaden the managers’ existing knowledge base (Katila & Ahuja, 2002; Levinthal & March, 1993; Mom, Van Den Bosch & Volberda, 2007). As frontline employees are directly confronted with new emerging technologies, changing customer demands and market conditions, they can be in possession of knowledge that is new and varies from managers’ existing knowledge base (Burgelman, 1983b; Mom et al., 2007; Ye, Marinova & Singh, 2012). In addition, according to Burgelman (1983b), knowledge that comes from lower level employees do not come about in a standardized or formalized way. These knowledge inflows are often random, impulsive and unplanned, and usually call for qualitative instead of quantitative changes of existing activities (Burgelman, 1983b; Sanchez & Heene, 1996, in Mom et al., 2007). This new and broad knowledge recombined with existing knowledge could trigger managers to modify existing beliefs, redefine strategic initiatives and to search for diverse novel solutions to occurring problems (Burgelman
1983b; Floyd & Lane, 2000; Kimberly, 1979, Quinn, 1985, in Mom et al., 2007). Hence, this paper suggests that bottom-up knowledge inflows will have a positive relation with managers’ exploration activities. This leads to the following hypothesis:
Hypothesis 1: Bottom-up knowledge inflows will be positively related to managers’ exploration activities
In order for managers to engage in exploitation activities, bottom-up knowledge inflows should deepen the existing knowledge base by adding knowledge that is closely related to the pre-existing knowledge base, so knowledge can be re-used to refine pre-existing certainties (Katila & Ahuja, 2002; March & Levinthal, 1993). Since bottom-up knowledge inflows are often ambiguous, unplanned and non-standardized, it is unlikely to yield knowledge that is closely related to a managers’ existing knowledge base. In addition, lower level employees are often not fully aware of the day-to-day activities for managers, and therefore could provide less new insights into how managers’ current activities might be refined or made more efficient. Hence, it is argued that bottom-up knowledge inflows will not deepen a managers’ existing knowledge base and therefore will not generate a positive impact on exploitation activities. Furthermore, this impact is also not expected to be negative. It could be argued that while bottom-up knowledge inflows trigger exploration activities, it might at the same time prevent managers from engaging in exploitation activities. However, prior research has indicated that exploration and exploitation activities are not mutually exclusive (Gibson & Birkenshaw, 2004), which means that individuals can engage in high levels of exploration while simultaneously engage in high levels of exploitation. Hence, triggering exploration activities will not downturn exploitation activities. It is suggested that bottom-up knowledge inflows are not related to managers’ exploitation activities.
2. Horizontal knowledge inflows and managers exploration and exploitation activities
Horizontal knowledge inflows refers to knowledge that comes from peer managers, either from the same unit or from other units within the same organization. Acquiring knowledge from peer managers is received in a different way than acquiring knowledge from below or above. While vertical knowledge flows are mostly based on asset specificity and achieving efficiency, horizontal knowledge flows brings in the importance of the social aspect of exchanging knowledge (Williamson, 1985). This type of knowledge is often transferred through personal relationships and informal networks, created by the rotation of employees among units (Minbaeva, 2010).
These interactions are commonly from different parts of the organization, which makes the knowledge often ambiguous, complex and tacit (Daft & Lengel, 1986; Egelhoff, 1991). Peer managers from different parts of the organization generally do not know a lot about managers’ current activities. For this reason, it is assumed it is harder for these types of knowledge inflows to be effective for analyzing, improving and refining the existing activities of a manager (Daft & Lengel, 1986, Egelhoff, 1991, in Mom et al., 2009). Furthermore, due to the ambiguity and complexity of these knowledge flows, it is unlikely to increase managers’ reliability in experience (Mom et al., 2007). Hence, it is suggested that horizontal knowledge inflows will not generate a positive impact on exploitation activities. It is furthermore expected that horizontal knowledge inflows will not prevent managers from engaging in exploitation, since exploration and exploitation activities are not mutually exclusive (Gibson & Birkenshaw, 2004). Managers are able to engage more in exploration activities, while not engaging less in exploitation activities. It is suggested that horizontal knowledge inflows will have no impact on managers’ exploitation activities.
On the contrary, a different impact of horizontal knowledge inflows on exploration activities is expected. Receiving horizontal knowledge flows through personal interactions, offers managers new knowledge that increases their variety in experience (Katila & Ahuja, 2002; Van Den Bosch & Van Wijk, 1999, in Mom et al., 2009). Through cross-functional interactions, new knowledge is recombined with managers’ existing knowledge and thereby boost innovative ideas (Grant, 1996;
Thompson, 1967, in Mom et al., 2009). In addition, because these knowledge flows are from different parts of the organization, it tends to be about unknown processes and activities, which as a result broadens a managers’ existing knowledge base (Winter & Szulanki, 2001). The recombined knowledge can trigger managers to modify existing beliefs, enhance innovative ideas and redefine strategic initiatives (Burgelman 1983b, Floyd and Lane, 2000, Kimberly, 1979, Quinn, 1985, in Mom et al., 2007). Hence, due to the fact that horizontal knowledge inflows enhances variety in experience and innovation, and broaden the managers’ existing knowledge base, the following hypothesis is suggested:
Hypothesis 2: Horizontal knowledge inflows will be positively related to managers’ exploration activities.
3. Top-down knowledge inflows and managers exploration and exploitation activities
Top-down knowledge inflows of a manager refers to knowledge that is acquired from a higher hierarchical level than the manager. In other words, this would be knowledge that a manager receives from its supervisor or employees at an even higher hierarchical level than its supervisor. Top-down knowledge inflows typically differ from bottom-up and horizontal knowledge flows.
For managers to engage in exploitation activities, they need knowledge that deepens their existing knowledge base in order to improve and refine their existing day-to-day activities (Katila & Ahuja, 2002). Both the knowledge recipient and the knowledge receiver, in this case the manager and the supervisor or someone at an even higher hierarchal level, are based in the same organizational unit and thus the knowledge is assumed to be about the area of expertise the manager is specialized in (Gupta & Govindarajan, 1991). Since this knowledge is related and even restricted to this area of expertise, it is expected to generate knowledge that is related to managers’ existing knowledge base (Winter & Szulanksi, 2001, in Mom et al., 2007). Hence, this knowledge is less new and diverse and likely to be more clear and understandable as they often come about in a planned and standardized way (Burgelman, 1983b; Sanchez & Heene, 1996, in Mom et al., 2007).
This type of knowledge helps managers to enhance, polish and refine their expertise in a specialized area (Schulz, 2001). In addition, due to the fact that these knowledge inflows are less ambiguous, it permits managers to develop their reliability in experience, instead of variance in experience (Daft & Lengel, 1986; Galunic & Rodan, 1998, in Mom et al., 2007). Furthermore, the top-down knowledge inflows relevance regarding refining managers’ existing activities is generally well-known, which as a result permits managers to respond to this information in acquainted ways. This will enhance the ability to efficiently implement current activities (Daft & Lengel, 1986; Galunic & Rodan, 1998, in Mom et al., 2007). For the reasons mentioned above, top-down knowledge inflows are suggested to have a positive impact on exploitation activities. Hence, this leads to the following hypothesis:
Hypothesis 3: Top-down knowledge inflows will be positively related to managers’ exploitation activities
When focusing on exploration activities, it is known that top-down knowledge inflows will not offer managers new, complex or varied knowledge that will broaden their existing knowledge base (Gupta & Govindarajan, 1994). However, senior management is able to generate an influence on managers’ exploration activities through other means. Senior management could trigger managers to engage in exploration activities by changing the structure of the organization. This could be done by decreasing formalization rules, increasing managers’ involvement in decision-making or implementing cross-functional meetings (Duncan, 1967; Egelhoff, 1991; Galbraith, 1973; McGrath, 2001; Tushman & O’Reilly, 1996, in Mom et al., 2007). In addition, senior or top management could also foster a culture that allows managers to deviate from their day-to-day work, challenge existing beliefs and provide strategic initiatives (Bartlett & Goshal, 1993, Volberda, 1998, O’Reilly and Tushman, 2004, in Mom et al., 2007). However, if the focus is purely on knowledge flows, it is suggested that top-down knowledge inflows will not have an impact on managers’ exploration activities (Mom et al., 2007).
4. The moderating effects of managers’ need for cognitive closure
The motivation to find closure has a large impact on the way individuals process the
knowledge that they receive (Kruglanski & Webster, 1996). Individuals with a high level of NFCC tend to make a shallow analysis of the acquired knowledge and are not motivated to look for knowledge that is inconsistent with their beliefs (Bukowski, Von Hecker & Kossowska, 2013). For knowledge to actually transfer, it depends on how easily knowledge sources are interpreted and absorbed (Kogut & Zander, 1992, in Van Wijk et al., 2008). Managers generally vary in their degree of knowledge acquisition and use, because of their individual differences in motivation to acquire and use the knowledge (Minbajeva et al., 2012). Previous research found that individuals who are extensively looking for and identifying knowledge show positive associations with the notion of using this knowledge for different activities (Tippmann, Managematin & Scott, 2013). As individuals with a high level of NFCC tend to make use of a shallow and narrow information search, they will probably make less use of particular acquired knowledge compared to individuals with a low level of NFCC,
especially when the knowledge flows are ambiguous and complex. This study suggests that
managers with a high level of NFCC absorb particular knowledge inflows differently than managers with a low level of NFCC due to difference in motivation. As a result this might impact the influence these knowledge flows have on managers’ exploration and exploitation activities.
4.1 Bottom-up and horizontal knowledge inflows and managers’ exploration activities: the moderating effect of managers’ need for cognitive closure
Bottom-up and horizontal knowledge inflows are expected to both have a positive impact on the exploration activities managers engage in. To discover the impact NFCC could have on these relationships, it is interesting to take a look at the contradictory results that are found (Mom et al., 2007, Wei et al., 2011). Regarding bottom-up knowledge flows, Mom et al. (2007) found that these knowledge inflows were positively related to the extent to which this manager engages in
exploration activities. In addition, Haynie (2009) found that managers have difficulties recognizing potential opportunities for exploration if they are not near the frontline or receive knowledge from frontline employees. On the contrary, Wei et al. (2011) found that bottom-up knowledge inflows have an inverted u-shaped effect on a manager’s exploration activities. Wei et al. (2011) proposes that the main reason why this declining effect occurs is because higher levels of new information could lead to increasing diverse and complex information which could be too difficult to integrate (Katila & Ahuja, 2002, in Wei et al., 2011). Frontline employees may offer different views on situations and if managers receive knowledge from a broad range of lower level employees, the scope of knowledge could get too diverse and complex. Too high levels of diversity and complexity of new knowledge could hurt exploration activities, mainly because the costs will be too high to integrate all this knowledge (Wei et al., 2011). These arguments seem appropriate, however managers could handle complex and diverse information differently. Managers with a high level of NFCC, for instance, will struggle more with new and complex information than managers with a low level of NFCC (Kruglanski & Webster, 1996). In addition, if processing the knowledge is perceived as effortful or costly, people tend to ignore this knowledge (Webster & Kruglanski, 1994). Receiving impulsive and varied bottom-up knowledge inflows might make managers feel negatively towards these knowledge inflows. It is therefore possible to suggest that it is not the complex and diverse information that declines the positive effect between bottom-up knowledge inflows and managers’ exploration activities, but the motivation from a manager to deal with this type of information and if it will actually be used to explore new opportunities. Hence, managers’ NFCC could have a weakening impact on the relation between bottom-up knowledge inflows and managers exploration activities. In other words, it is assumed that managers with a high level of NFCC will use bottom-up knowledge inflows less for their exploration activities, compared to managers with a low level of NFCC.
Regarding horizontal knowledge inflows, the characteristics of these knowledge inflows are more or less similar to the characteristics of bottom-up knowledge inflows, in the sense of being
broad and often impulsive, and so the arguments for a weakening effect of NFCC are roughly the same. A profound difference from these two types of knowledge inflows is that horizontal knowledge inflows tend to be acquired from different parts of the organization and as a result are more vague and ambiguous than bottom-up knowledge inflows. NFCC is negatively associated with ‘tolerance for ambiguity’ (Webster & Kruglanski, 1994). Having a high tolerance for ambiguity generates the ability to deal effectively with vague, unclear, unstructured and incomplete information (Budner, 1962, in Gupta & Govindarajan, 2000). Hence, managers with a high NFCC and a low tolerance for ambiguity might find it difficult to deal with the vague and unknown knowledge they receive. As a result, their motivation for using the received knowledge will be lower. In addition, it is easier to ‘ignore’ information from managers from different units, as they will notice less what has actually been done with the acquired knowledge. Due to the several arguments mentioned above, this paper suggests that managers with a higher level of NFCC will use horizontal knowledge inflows less for exploration activities than managers with a lower level of NFCC. The following hypotheses concerning bottom-up and horizontal knowledge, and the motivation to use these knowledge flows for exploration activities, are proposed:
Hypothesis 4a: Managers’ need for cognitive closure will weaken the relationship between bottom-up knowledge inflows and managers’ exploration activities
Hypothesis 4b: Managers’ need for cognitive closure will weaken the relationship between horizontal knowledge inflows and managers’ exploration activities
4.2 Top-down knowledge inflows and managers’ exploitation activities: the moderating effect of managers’ need for cognitive closure
Since top-down knowledge flows are expected to be specialized, clear, understandable and more related to a managers’ existing knowledge base, it is assumed that NFCC will have a different impact compared to bottom-up and horizontal knowledge flows. Top-down knowledge flows deepen
a managers existing knowledge base and is therefore all about “the use and development of things already known” (Levinthal & March, 1993, p. 105). Hence, this knowledge is generally less new, diverse, ambiguous and complex than knowledge that is used for exploration activities and as a result the perceived effort and costs will be lower. It will thus be less likely that the information is being ignored or not processed (Kruglanksi & Webster, 1994) due to a high level of NFCC. On top of that, knowledge that comes from a manager’s supervisor or from someone with an even higher position will probably be dealt with differently than knowledge that comes from lower level employees. If, for example, someone at a higher hierarchical position offers information on how to improve or refine existing activities, the manager will probably not just ignore this information. Potential reasons for this could be that they are frightened to be criticized, lose promotion opportunities or lose their job. Ignoring information from lower level employees or managers from different units is assumed to be much easier as those employees generally do not have the authority to ‘punish’ managers when their knowledge inflows are being disregarded (Paswan, Peltan & True, 2005). In addition, knowledge inflows of a manager’s supervisor could be direct instructions or orders. For the reasons mentioned above, top-down knowledge inflows will generally be listened to for exploitation activities, even in the case of a high level of NFCC.
However, a high level of NFCC might even enhance the relation between top-down knowledge inflows and exploitation activities. As mentioned before, NFCC is the desire to seize on closure quickly and freeze on it so no additional data is needed (Kruglanksi & Webster, 1996). When high-level NFCC managers receive information regarding exploitation activities, they might want to use this information so a solution to a problem is being solved more quickly and they feel closure. The difference with exploration activities is that these activities are less necessary and give less pressure for the short term (March, 1991). As a result, managers with high NFCC might want to find closure on a problem concerning exploitation activities faster than concerning exploration activities. Managers with low levels of NFCC possibly want to wait with using the received knowledge to conduct a broad and diverse knowledge search on the topic through different sources. It is therefore
assumed that low NFCC managers will feel less motivation to find closure immediately. For the reasons mentioned above, this paper suggests that managers with a high level of NFCC will use top-down knowledge inflows more for exploitation activities, than managers with a low level of NFCC. Hence, this leads to the following hypothesis:
Hypothesis 4c: Managers’ need for cognitive closure will strengthen the relationship between horizontal knowledge inflows and managers exploration activities
5. Conceptual model
The six hypotheses that are outlined in the theoretical framework are showcased in the following conceptual model:
Managers’ need for cognitive closure Bottom-up knowledge inflows Horizontal knowledge inflows Top-down knowledge inflows Exploration activities Exploitation activities H1 H2 H3 H4a H4b H4c Control variables: - Gender - Age - Education - Tenure within ABN
AMRO
- Tenure within current function
IV Methodology
The following paragraphs will discuss the main information about the research approach of this paper. First of all, the research design will be portrayed. Second, the sample and data collection will be presented and third, the measurements of the independent variables, dependent variables, moderator and control variables will be depicted.
1. Research design
The research that is conducted for this paper could be characterized as confirmatory research, as beforehand made hypotheses are being tested. In order to test hypotheses, quantitative data has been collected through the use of online questionnaires. Using questionnaires to collect data is often suggested because it can engender results at a smaller cost than gathering data for the whole population and it is relatively easy to compare responses (Saunders & Lewis, 2012). The use of online surveys generates a few important advantages. It allows access to participants in different locations, creates the ability to reach respondents that are difficult to contact and makes use of automatic data collection so time and effort is reduced (Wright, 2005). The online survey was appropriate for this research, as a large sample of middle level managers needed to be approached, and transmitting the survey by email was a simple and effective way to reach this sample group. In addition, the amount of responses for emailing surveys tend to be equal or better than surveys that are distributed traditionally (Metha & Sivadas, 1995; Stanton, 1998; Thompson, Surface, Martin & Sanders, 2003, in Wright, 2005).
A possible downside of using online questionnaires is that the participants need to be computer-literate individuals (Saunders et al, 2009). However, nowadays, almost every company uses Internet and email for communication and information (Fox et al., 2001; Nie et al., 2002, in Wright, 2005), hence it is unlikely that the respondents did not understand the way the survey should be filled out.
2. Sample and data collection
For this research , a large sample of middle level managers was needed, preferably in the same company so biases for organizational cultures and different industries could be minimalized. Moreover, it had to be a large company with enough hierarchal levels, as well as a company where employees feel the pressure to both explore and exploit. The latter criterion will enhance the possibility of measuring managers’ exploration and exploitation activities (Mom, Van Den Bosch & Volberda, 2007).
In the financial services industry, changes regarding costumer demands, competition and new technologies are pushing companies to explore (Banker et al., 2005; Flier et al., 2001; Gibson & Birkenshaw, 2004, in Mom et al., 2007). However, simultaneously, financial services companies need to focus on the efficiency of their day-to-day activities due to short term pressures, and thus are also being pushed to exploit. Hence, approaching companies within this industry seemed like an excellent choice for this research. ABN AMRO, one of the largest banks in the Netherlands, was willing to participate with the study, due to the practical implications the results of this research could offer them.
ABN AMRO serves 5.5 million customers and employs 5500 people. The organization consists of six major departments: (1) Finance, (2) People regulations and identity, (3) Risk management and strategy, (4) Retail and private banking, (5) Commercial and merchant banking, and (6) TOPS. The last department, ABN AMRO TOPS, seemed to fit the most with the requirements for this research. This business line is sometimes referred to as the ‘back bone’ of the company as all operational processes are being facilitated from here. One could say that employees within the TOPS department are engaging extensively in exploitation activities. However, ABN AMRO TOPS is aiming to become a pro-active innovation platform and wants their employees to strive for new ideas that make a difference for their clients. Hence, besides working on making the usual tasks more efficient – exploitation – employees are being motivated to discover and experiment with new opportunities – exploring. In addition, due to the many hierarchal levels within ABN AMRO, managers are able to
receive knowledge from different layers, which made it easier to measure knowledge inflows. For the reasons mentioned above, middle level managers within ABN AMRO TOPS were assumed to be appropriate participants for this research.
In collaboration with the HR Director of ABN AMRO TOPS, this research was conducted at six disciplines – IT Services and Solutions, Facility management, Operations, COO C&MB, COO R&PB en COO Functions – of the TOPS department within ABN AMRO. In these disciplines 277 middle level managers were approached. All respondents were operating on the same hierarchical level and suited the middle level manager criteria (Huy, 2002), so bias due to the sampling procedure may not be a problem.
The HR director of ABN AMRO TOPS has spread the online survey by email (see appendix 1), after getting approval from the six executives of the six disciplines within TOPS. To encourage the middle level managers to fill out the survey, the HR director of ABN AMRO TOPS distributed the email via her account. The email mentioned that the results of the survey are used for academic purposes only, but ABN AMRO and the managers themselves could receive the results in order to get more insights in the activities middle level managers engage in. The email included a document with information on the research and a link to the online survey (see appendix 2). The online survey consisted of six pages with the last page devoted to demographic measures. To guarantee confidentiality, the names of the respondents were not revealed. To increase the response rate, a reminder email was sent after two weeks. In the end, 100 middle level managers took the effort to fill out the survey, which indicates a response rate of 36.1%.
3. Measurements
In the following paragraphs the measurements of the main variables – exploration,
exploitation, bottom-up , horizontal , top-down knowledge inflows, and need for cognitive closure – and the control variables will be presented. All scales were adopted from previous research, to ensure well-tested and valid scales. To make sure the respondents would perfectly understand the
survey, all questions were translated to Dutch. The translation was checked by several people to make sure the validity of the scales would not change.
3.1 Dependent variables: managers’ exploration and exploitation activities
The exploration and exploitation scales were adopted from Mom et al. (2007, 2009). As there used to be no appropriate scales for exploration and exploitation on an individual level of analysis yet, Mom et al. (2007, 2009) developed items based on the characteristics for exploration and exploitation (March, 1991). Respondents were asked to answer questions about how much they were involved in exploration and exploitation activities last year.
The exploration and exploitation activities were both measured with seven items on a 7-point scale (1 = to a very small extent to 7 = to a very large extent). The items are showcased in table 1 and table 2 below.
Table 1. Items exploration activities
To what extent did you, last year, engage in work related activities that can be characterized as follows:
1 Searching for new possibilities with respect to products/ services, processes or markets
2 Evaluating diverse options with respect to products/ services, processes or markets 3 Focusing on strong renewal of products/ services or processes
4 Activities of which the associated yields or costs are currently unclear 5 Activities requiring much adaptability from your side
6 Activities requiring you to learn new skills
Table 2. Items exploitation activities
The reliability for both scales has been checked. Reliability indicates whether a construct is stable and consistent. According to Field (2009), constructs could be perceived as sufficiently reliable when the Cronbach’s alpha exceeds 0.70. Both exploration and exploitation are reliable: exploitation α = 0.840; exploration α = 0.721.
To assess whether the constructs measured what they should measure, the discriminant and convergent validity for exploration and exploitation have been checked. Convergent validity explores if it is likely that there is one underlying factor that might have had an effect on all item scores or that there are multiple underlying factors. Discriminant validity assesses whether the items really belong to the construct that they are supposed to reflect and do not fit better with some other construct (Campbell & Fiske, 1959). Exploratory factor analysis demonstrates that the fourteen items for exploration and exploitation load on four components that have an eigenvalue higher than 1. The exploitation items all load on one component, while the exploration items load on two components. (see appendix 3, table 12). There is one item for exploration and one item for exploitation that also load on a fourth component. Factor analysis showed that for exploitation the most important underlying factor explains approximately 58 % of the variance and thus confirms the convergent validity of the construct. Since the exploration construct loads on two factors, deleting exploration items 5 and 6 could create a better convergent validity. However, when measuring the reliability from the exploration construct after the two items are deleted, it reveals that the reliability To what extent did you, last year, engage in work related activities that can be characterized as follows:
1 Activities of which a lot of experience has been accumulated by yourself 2 Activities which you carry out as if it were routine
3 Activities which serve existing (internal) customers with existing services/ products 4 Activities of which it is clear to you how to conduct them
5 Activities primarily focused on achieving short-term goals
6 Activities which you can properly conduct by using your present skills 7 Activities which clearly fit into existing company policy
decreases from a Cronbach’s alpha of .721 to a Cronbach’s alpha of .682. Since, the reliability coefficients were indicating a good correlation with the rest of the scale and are contributing to the overall Cronbach’s alpha value, no items were deleted in order to create a better convergent validity for exploration. In addition, the exploration and exploitation scale has been extensively tested and considered valid in previous research (Mom et al., 2007). In this study factor analysis showed that for exploration the two most important underlying factors explain 38% and 21% of the variance. Table 12 (appendix 3) does show that the items for exploration and exploitation do not load on each other’s factors. It could thus be argued that the two construct have a good discriminant validity.
3.2 Independent variables: bottom-up, horizontal and top-down knowledge inflows
The knowledge inflows are also measured with a scale conducted by Mom, Van Den Bosch and Volberda (2007). The items were measured on a 7-point scale (1 = to a very small extent to 7 = to a very large extent). When filling out the survey, participants were asked to think of explicit and tacit knowledge that concerns processes, systems, technologies, strategies, products and markets. In addition, they were requested to consider different knowledge channels through which the knowledge can flow such as face-to-face contacts, email, telephone, (in)formal meetings and shared technologies (Mom, Van Den Bosch & Volberda, 2007).
Table 3. Items knowledge inflows
Factor analysis confirms the convergent validity for the independent variables top-down knowledge inflows, bottom-up knowledge inflows and horizontal knowledge inflows by exposing that each independent variable’s items load on one factor only. The most important underlying factor explains approximately 67 %, 60 % and 65% of the variance respectively. To check for discriminative validity, factor analysis has been performed with all the items that are supposed to measure bottom-up knowledge inflows, top-down knowledge inflows and horizontal knowledge inflows. Results of the factor analysis showed that there are three components with an eigenvalue higher than 1, and that the components reflect the three different independent variables (see table 13, appendix 4).
Furthermore, the reliability for the three independent variables has been checked. Both horizontal knowledge inflows (α = .723) and top-down knowledge inflows are reliable (α = .751). However, the independent variable bottom-up knowledge inflows could not be seen as a reliable construct, α = .320. The construct bottom-up knowledge inflows consists of two items: To what extent did you, last year, receive or gather knowledge from (1) your direct assistants, (2) on more To what extent did you, last year, receive or gather knowledge from:
Top-down knowledge inflows 1 Your direct supervisor
2 One more hierarchical level up than your direct supervisor 3 Two more hierarchical levels up than your direct supervisor Horizontal knowledge inflows
1 Peer teams within your own organizational unit
2 Peer teams in other organizational units within your own division 3 Peer teams in other divisions
Bottom-up knowledge inflows 1 Your direct assistants
hierarchical level down than your direct assistants?. Since middle level managers in general will receive more knowledge from their direct assistants, looking at the mean of the two items is not appropriate in this case. Therefore bottom-up knowledge inflows is created as a construct in SPSS that sums up the two items.
3.3 Moderating variable: managers’ need for cognitive closure
The original need for closure scale was developed as a one-dimensional construct (Webster & Kruglanksi, 1994). However, Neuberg, Judice and West (1997) came up with a two dimensional need for closure construct consisting of items from the facets predictability, order, ambiguity, and close-mindedness and items from the facet decisiveness. There has been a debate going on about whether the decisiveness items measures the ability to achieve cognitive closure or the motivation and need to achieve cognitive closure. As need for closure should be a concept that measures the motivation to accomplish cognitive closure, six alternative decisiveness items were constructed and replaced the seven original decisiveness items (Roets & Van Hiel, 2007). The altered 41-item need for cognitive closure scale is now being recognized as “the improved version of the scale” (Kruglanksi, Dechesne, Orehek & Pierro, 2009, p. 151, in Roets & Van Hiel, 2011).
The 41-item need for closure scale is used a lot in the field of psychology and business, but the length of the scale is considered a problem. Multiple researchers therefore used a shortened version by selecting items on their own. These items were often not carefully chosen on appropriate grounds, so Roets and Van Hiel (2011) decided to design an empirically validated reduced version of the need for closure scale consisting of fifteen items. For practical reasons and because the 15-item scale is a valid alternative measure of need for cognitive closure, I have used the shortened version of the NFCC scale conducted by Roets and Van Hiel (2011) to measure managers’ NFCC. The items were measured on a 7-point scale (1 = strongly disagree to 7 = strongly agree). In the table below, the fifteen items for need for cognitive closure are showcased. Moreover, the moderating variable is reliable with a Cronbach’s alpha of α = .79.
Table 4. Items need for cognitive closure scale
3.4 Control variables
To be sure to study the right effects between the main variables, a few control variables needed to be tested for. Control variables can sort out the extraneous effect they have on relationships and thus increase the validity of the results (Saunders & Lewis, 2012). As this research was conducted within one department of one company with managers based on the same hierarchal level, fewer variables had to be controlled for. The managers are exposed to the same environment and the same level of tasks. However, the respondents could differ in demographics and personal characteristics that could have an effect on the engaging in exploration and/or exploitation activities. The control variables that were added in this research are: (1) gender, (2) age, (3) education, (4) tenure within ABN AMRO, (5) tenure in current function and (6) TOPS discipline. The following paragraph will explain the reasons for including these control variables.
To what extent do you agree with the following statements? 1 I don’t like situations that are uncertain
2 I dislike questions which could be answered in many different ways 3 I find that a well ordered life with regular hours suits my temperament
4 I feel uncomfortable when I don’t understand the reason why an event occurred in my life 5 I feel irritated when one person disagrees with what everyone else in a group believes 6 I don’t like to go into a situation without knowing what I can expect from it
7 When I have made a decision, I feel relieved
8 I would quickly become impatient and irritated if I would not find a solution to a problem immediately.
9 I don’t like to be with people who are capable of unexpected actions 10 I dislike it when a person’s statement could mean different things 11 I find that establishing a consistent routine helps me enjoy life more 12 I enjoy having a clear and structured mode of life
13 I do not usually consult many different opinions before forming my own view 14 I dislike unpredictable situations
