The effect of time pressure on individual ambidexterity : a vignette study

(1)

Master Thesis

The Effect of Time Pressure on Individual

Ambidexterity: A Vignette Study

Student:

David Ahlers – Student

№

11385316

MSc. Business Administration – Strategy Track

University of Amsterdam, Amsterdam Business School

Supervisor:

Dhr. MSc. B. Silveira Barbosa Correia Lima

Date:

23

rd

(2)

Statement of originality

This document is written by student David Ahlers 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.

Amsterdam, 23

rd

June 2017

David Ahlers

(3)

Abstract

Prior research regarding the concept of ambidexterity has largely focused at the organizational level. This study aims at refining ambidexterity at the individual level. More specifically, it examines the relationship between time pressure and individual ambidexterity. The paper develops and tests hypotheses on the effect of time pressure as well as the moderating role of resilience and openness to experience concerning the focal relationship between time pressure and individual ambidexterity. Based on the literature review, it is hypothesized that time pressure negatively influences the extent to which a person behaves ambidextrously. Concerning the exact direction of this effect, it is further assumed that people tend to focus more on exploitative than explorative activities when facing time pressure. In addition, it is predicted that both resilience and openness to experience negatively moderate the focal relationship, so that the effect of time pressure is mitigated for high levels of each characteristic. The results for this study are derived from the outcome of a conducted vignette study with 135 participants. By applying an experimental vignette design for the data collection, the study reacts on scientists’ calls to test the ambidexterity construct within dynamic settings. The final results provide support for the hypotheses related to the effect of time pressure. A significantly negative influence of time pressure on individual ambidexterity is demonstrated as participants tend to focus on exploitative tasks when working under time pressure. Surprisingly, the data analysis further reveals that neither resilience nor openness to experience significantly moderate the relationship between time pressure and individual ambidexterity.

Keywords: individual ambidexterity, time pressure, resilience, openness to experience, vignette study

(5)

1. Introduction

Organizations continuously face the challenge of ensuring short-term profitability while simultaneously sustaining success in the long run. Thus, they have to deal with the tension of exploiting their current capabilities and exploring new ones at the same time (March 1991, Levinthal & March 1993). More specifically, exploitation is based on existing knowledge and thus associated with activities such as efficiency and refinement. In contrast, exploration refers to notions that are based on new knowledge such as experimentation and discovery (March 1991). Hence, both forms of activities demand for opposing task orientations. The ability to balance the conflicting demands of exploitation and exploration is known as ambidexterity (Raisch et al. 2009).

Especially during the last two decades, the concept of ambidexterity has gained large attention in the literature (O’Reilly & Tushman 2013). Whereas earlier studies on ambidexterity often regarded the trade-off between exploration and exploitation as insurmountable (McGill et al. 1992), more recent research found concrete solutions to support ambidexterity within an organization (Tushman & O'Reilly 1996, Birkinshaw & Gibson 2004). So far, most of the scholars conducting ambidexterity research have focused at the organizational level. In order to gain a deeper understanding on how organizations can achieve ambidexterity, these scholars identified structural ambidexterity and contextual ambidexterity as two possible pathways for an organization to become ambidextrous (Duncan 1976, Tushman & O'Reilly 1996, Birkinshaw & Gibson 2004). Considering structural ambidexterity, organizations can either separate their exploitative and explorative activities temporally (Duncan 1976) or spatially (Tushman & O'Reilly 1996). On the other hand, the concept of contextual ambidexterity focuses on creating an organizational environment in which individuals are likely to engage in both, exploitative and explorative activities (Birkinshaw & Gibson 2004). Especially the contextual form of ambidexterity, which highlights the benefits of combining both activities in individual work roles, makes clear that the concept of ambidexterity consists of multiple levels – not just the

(6)

organizational. Ambidexterity can also be present at the group level or the individual level (Turner et al. 2013). Raisch and Birkinshaw (2008) underline the importance of focusing ambidexterity research on different levels of analysis. According to them, tensions at one level of analysis can often be resolved by taking a deeper look at a lower level of analysis. In other words, tensions at the organizational level may be resolved by investigating ambidexterity at the individual level of analysis.

Although the individual level of analysis represents an essential part in the ambidexterity construct, research at the individual level of ambidexterity is surprisingly limited so far (Good & Michel 2013). As a result, ambidexterity research still lacks an empirically and conceptually validated understanding of individual ambidexterity (Mom et al. 2007, 2009). Existing studies concerning ambidexterity at the individual level predominantly identify antecedents of individual ambidexterity and point out its crucial role within an organization. Most of them primarily focus on the ambidexterity of managers as well as the effects their behavioral characteristics and leadership style have on the ambidexterity of employees. Accordingly, Mom et al. (2007, 2009) and Rogan & Mors (2014) highlight the influence of managers’ knowledge inflows as well as the importance of their networks for ambidextrous behavior at the managerial level. Furthermore, scholars suggest that managers can positively influence employees’ ambidextrous behavior by combining strong managerial support with high performance expectations – also known as paradoxical leadership (Birkinshaw & Gibson 2004, Rosing et al. 2011, Kauppila & Tempelaar 2016). Besides this managerial perspective, a few more papers on individual ambidexterity examine the cognitive abilities that individuals need for balancing exploitative and explorative activities (Good & Michel 2013, Kauppila & Tempelaar 2016). Good and Michel (2013) show that certain cognitive abilities, such as focused attention or intelligence, can positively influence individual ambidexterity. Furthermore, Kauppila & Tempelaar (2016) demonstrate that general self-efficacy enables ambidextrous behavior.

(7)

Nevertheless, when considering prior research on individual ambidexterity, it becomes obvious that there is a scarcity of empirical data on certain essential research directions such as the influence of contextual features on individual ambidexterity (Good & Michel 2013, Birkinshaw & Gibson 2004). Gibson and Birkinshaw’s (2004) construct of contextual ambidexterity clearly demonstrates that the organizational context directly influences the degree to which individuals within that organization behave ambidextrously. In their study, they prove the direct relation between contextual factors and individual ambidexterity by showing that an organizational context, in which social support and performance management are well balanced, fosters individual ambidexterity. Although Gibson and Birkinshaw’s (2004) paper raised attention to a well-balanced organizational context as an important antecedent for individual ambidexterity, there are still several contextual factors which have not been taken into account. More concretely, Good and Michel (2013) call for future research on the influence of time pressure as a contextual factor of individual ambidexterity. Yet, what makes especially time pressure such a noteworthy contextual factor to explore in relation to individual ambidexterity?

When reviewing the literature on time pressure in work environments, it can be observed that time pressure has a direct influence on people’s work performance. While most of the scholars agree on the existing relationship between time pressure and performance, there is no clear consensus on the direction of influence (Hsiao et al. 2015). Whereas several research projects indicate that work performance and time pressure are positively related (Kelly & Karau 1993, Latham & Locke 1975), other scholars show a negative relation (Simonton 1999, Locke & Whiting 1974). Andrews and Faris (1972) even add a third category by stating that work performance and time pressure have a reverse U-shaped relationship. Relating time pressure to individual ambidexterity, one potential scenario is that individuals will focus more on activities that are based on existing knowledge than on developing new knowledge when facing time pressure. Thus, people would favor exploitative activities over explorative (Beck & Schmidt

(8)

2013). However, existing research on time pressure suggests that this is only one possible way. Even though none of these existing studies examines individual ambidexterity, several research papers indicate contradicting results on the effect of time pressure on explorative work behavior (Hsiao et al. 2015, Blank et al. 2014, Amabile et al. 2002, Beck & Schmidt 2013). Whereas Beck and Schmidt (2013) and Amabile et al. (2002) point out the largely negative effects of time pressure on creative cognitive processes and on developing new skills, Hsiao et al. (2015) and Blank et al. (2014) reveal a positive influence of time pressure on creativity and innovation. Accordingly, time pressure can be identified as an important contextual factor with a direct but ambiguous impact on individual ambidexterity.

The majority of prior studies on individual ambidexterity fail to capture the balance between explorative and exploitative activities in dynamic settings. This lack of research may be caused by the difficulty of collecting individual data in a dynamic context. The static approaches in which participants are asked about their former behavior concerning exploring and exploiting can indeed identify contextual factors that foster individual ambidexterity. Nonetheless, those approaches do not consider varying dynamic settings such as different levels of time pressure. Thereby, these studies do not measure in which situations the level of individual ambidexterity is higher than in others since they typically only observe the average degree of ambidexterity carried out over a longer period (Good & Michel 2013). Taking all together, researching the relationship between time pressure and individual ambidexterity represents an important gap in the literature of individual ambidexterity that needs to be examined. Consequently, the research question that will be investigated in this paper is:

What influence does time pressure have on individual ambidexterity?

The dynamic setting of the study is achieved through an experimental vignette design, in which participants are asked to rank explorative and exploitative tasks while facing different levels of time constraint. The study defines individual ambidexterity as a construct in which an

(9)

individual pursues both explorative and exploitative activities in a balanced way. It describes an individual’s behavioral capacity to engage in and combine opposing task performances (Kauppila & Tempelaar 2016, Bledow et al. 2009). In accordance with the balance dimension of ambidexterity, individual ambidexterity reaches high levels when the absolute difference of exploration and exploitation is low (He & Wong 2004, Cao et al. 2009).

In addition to the examined main relationship between time pressure and individual ambidexterity, it is important to consider individual characteristics of each participant as they might account for differences in work behavior (Gupta et al. 2006, Raisch et al. 2009, Good & Michel 2013, Kauppila & Tempelaar 2016). Therefore, the study tests the moderating effect of resilience and openness to experience on the focal relationship between time pressure and individual ambidexterity. Resilience is defined as a process whereby people exposed to high pressure, demands and environmental difficulties are able to use positive mental skills to remain psychologically steady and focused (Ong et al. 2006, Kinman & Grant 2011, Juster & Marin 2013, Shatté et al. 2017). Thus, it will be interesting to observe if individuals who possess a high level of resilience are able to cope better with time pressure and show a different, possibly more ambidextrous, work behavior than those with a lower level (Shatté et al. 2017). People who are open to experience can be described as imaginative, broad-minded, curious and perceptive (Barrick & Mount 1991, Barrick et al. 2001). Several studies on openness to experience suggest that this characteristic can influence the balance of explorative and exploitative behavior – mostly in a positive way (Lepine et al. 2000, Keller & Weibler 2015). For this research, it will be especially interesting to observe if the influence of this personality trait is significant enough to moderate the relationship between time pressure and individual ambidexterity and if so, how it will affect the balance between exploration and exploitation.

By investigating the influence of time pressure on individual ambidexterity, the study makes contributions to both theory and practice. First of all, it contributes to the current body of knowledge on individual ambidexterity by adding time pressure as a relevant contextual

(10)

factor that might positively or negatively affect an individual’s ambidexterity. Thereby, the paper directly responds to scientists’ call for a better understanding of the role of context within the ambidexterity research field (Gibson & Birkinshaw 2004, Lavie et al. 2010, Rausch et al. 2009, Good & Michel 2013). Since this research tests individual ambidexterity in an experimental setting, it belongs to the very few studies within ambidexterity research that investigate the construct in a dynamic context (Good & Michel 2013). By considering the direct impact of time pressure, the study is able to measure whether time pressure enhances or diminishes high levels of individual ambidexterity. In addition, it identifies if people tend to prioritize explorative or exploitative activities when they face an increasing time constraint. Furthermore, the analysis of the moderating role of resilience and openness to experience sheds light on the individual differences in work behavior among people when facing a certain level of time pressure. It is an important point to examine the exact influence of time pressure on individual ambidexterity, yet another one to identify levers in form of individual characteristics that can explain deviations among individuals. The research expands prior studies that investigated the direct impact of a characteristic on individual ambidexterity since it measures whether the respective personality traits enable individuals to better handle the exploration-exploitation tension under time pressure. In practice, these results can help managers to either capitalize on the beneficial impact of time pressure or to minimize its detrimental consequences for individual ambidexterity. Understanding the direct effect of time pressure on employees’ ambidextrous behavior can transform time pressure into a lever in such way that it can either be increased or decreased by managers, depending on the desired result.

(11)

2. Literature review

The following chapter summarizes and discusses the most relevant insights of prior research papers related to the investigated topic of this study. Taken together, the combination of these insights provides the theoretical framework of the paper from which the hypotheses for this research are derived. The first section conceptualizes exploration and exploitation since these terms present the fundamental basis for every ambidexterity research. Next, the most relevant findings on individual ambidexterity are discussed. Subsequently, the concept of time pressure is introduced and directly linked to individual ambidexterity which leads to the first two hypotheses. Finally, the moderating role of resilience and openness to experience is examined by relating both personality traits to time pressure and individual ambidexterity. This section entails the remaining two hypotheses of the study.

2.1 Exploration vs. exploitation

The concepts of exploration and exploitation have been studied in many different contexts within existing management literature. Scholars in the research fields of organizational learning (Levinthal & March 1993, March 1991), organizational design (Tushman & O’Reilly 1996), knowledge management (Brown & Duguid 2001) as well as adaptation (Eisenhardt & Brown 1997) respectively built a conceptual understanding of the exploration-exploitation notion. According to the varying contexts in which these terms have been investigated, there are many different definitions for the concepts of exploration and exploitation (Lavie et al. 2010). As both concepts build the essential basis for every ambidexterity research, it is important to determine the definition of exploration and exploitation that will be used in this study.

In his research paper on exploration and exploitation in organizational learning, March (1991) defines exploration as a concept that comprises activities which are primarily based on new knowledge such as experimentation, flexibility, discovery and innovation. On the other hand, he relates exploitation to activities that build on existing knowledge like refinement,

(12)

choice, efficiency and execution. Furthermore, he underlines that both concepts are equally fundamental for the long-term survival of an organization. Consequently, an organization that engages in exploration while neglecting exploitation will most likely suffer the costs of innovating or experimenting without benefiting from the advantages. On the other hand, a complete focus on exploitation to the exclusion of exploration might initially lead to an achievement of short-term goals, but will prevent the organization from keeping pace with competitors in the long run. March (1991) concludes that „maintaining an appropriate balance between exploration and exploitation is a primary factor in system survival and prosperity” (p.71).

Nevertheless, this is not always easy because both explorative and exploitative activities typically compete for scarce resources within an organization. Tushman and O’Reilly (1996) elaborate on the suggested difficulty of balancing exploration and exploitation by stating that both concepts follow a fundamentally different logic and thus demand for opposing strategies and structures that are complicated to reconcile. Gupta et al. (2006) scrutinize the approach of both studies by investigating the question whether exploration and exploitation are two ends of a continuum or rather independent aspects of organizational behavior. Their results indicate that the scarcity of resources that are needed to pursue both exploration and exploitation indeed makes the two activities mutually exclusive. That means, high values of either exploration or exploitation will consequently imply low values of the other. Moreover, they claim that exploration and exploitation will always be mutually exclusive within a single domain, such as an individual person. Nonetheless, they also discover that exploration and exploitation are rather orthogonal across different and loosely coupled domains. That is, high levels of either exploration or exploitation in one domain can coexist with high levels of the opposing activity in another domain. As a result, they are convinced that an all-embracing argument in favor of either continuity or orthogonality cannot be made. In fact, the relationship between exploration

(13)

and exploitation depends on the scarcity of resources and whether these activities take place within a single domain or across multiple domains.

Following the presented findings in the literature, this study defines exploration and exploitation as two ends of a continuum whenever the focal unit of analysis is a single domain, e.g. an individual. This means, that within this domain, a balance between exploration and exploitation has to be achieved. In line with Gupta et al.’s (2006) assumptions, an individual is often facing scarce resources, such as a time famine, when executing tasks. Since exploration and exploitation are defined as two ends of a continuum, individuals have to decide if they rather focus on exploitative or explorative activities or balance both equally when time is limited. The concept of exploration itself represents activities that are based on gaining new knowledge. This new knowledge can be gained through actions like experimenting and creative or innovative thinking. In contrast, exploitation is defined as a concept comprising activities that build on existing knowledge. Examples for exploitative behavior are efficiency and refinement of current processes.

2.2 Individual ambidexterity

Whereas the previous subchapter clarified the underlying concepts of the ambidexterity phenomenon, this subchapter focuses on how the balance of both exploration and exploitation actually leads to ambidexterity – at the individual level.

Individual ambidexterity is defined as a construct that indicates the extent to which an individual pursues both explorative and exploitative activities in a balanced way. It describes an individual’s behavioral capacity to engage in and combine opposing task performances (Kauppila & Tempelaar 2016, Bledow et al. 2009). In accordance with the balance view of ambidexterity, the individual ambidexterity of people reaches high levels when the absolute difference of exploration and exploitation is low. In line with the conceptualization made in the

(14)

previous chapter, this view considers the relative magnitude of an individual’s explorative and exploitative activities (He & Wong 2004, Cao et al. 2009).

Although most of the existing literature in the ambidexterity research field focuses on ambidexterity at the organizational level, ambidexterity is a multilevel construct (Good & Michel 2013). Hence, ambidexterity within an organization is not simply driven by a certain strategy to engage in both exploration and exploitation, but also by the business units (Gibson & Birkinshaw 2004) and individuals of the organization (Mom et al. 2007, 2009). Whereas earlier studies on ambidexterity often regarded the trade-off between exploration and exploitation as insurmountable (McGill et al. 1992), more recent research found different pathways at multiple levels of analysis to achieve ambidexterity within an organization (Tushman & O'Reilly 1996, Birkinshaw & Gibson 2004). This study is focusing on the pathway of individual ambidexterity. This is not primarily because it represents a field within the ambidexterity research that is still not sufficiently explored, but rather because of the importance of the individual level for the overall understanding of the ambidexterity phenomenon. While the organizational level of ambidexterity is a more visible construct that can be directly linked to the organization’s output, the individual level of ambidexterity constitutes a yet less understood sublevel in which many organizational problems are rooted and can be solved (Raisch & Birkinshaw 2008). Thus, a deeper understanding of the individual level can lead to a better perception of ambidexterity at the organizational level, which ultimately drives firm profitability (Tushman & O'Reilly 1996, He & Wong 2004, Gibson & Birkinshaw 2004, Lubatkin et al. 2006, Auh & Menguc 2005).

While most of the researchers in the ambidexterity field aim at refining the already extensive knowledge of ambidexterity at the organizational level, Gibson and Birkinshaw (2004) were the first who elaborated on the individual level. By introducing the concept of contextual ambidexterity, they link organizational success with the ability to create an appropriate organizational context that encourages employees to engage in both explorative and

(15)

exploitative activities. The approach is based on the assumption that exploration and exploitation can be maintained simultaneously at every organizational level when individuals within the organization are able to balance explorative and exploitative activities themselves (Lavie et al. 2010). According to Ghoshal and Bartlett (1994), the degree of four sets of attributes - support, trust, discipline and stretch - determine an organizational context. Gibson and Birkinshaw (2004) argue that an organizational culture that combines all four attributes in a balanced way will empower individuals to maintain a balance between opposing task orientations like creativity and efficiency. The combination of discipline and stretch is defined as performance management, the combination of support and trust as social support. Thus, in an organization that emphasizes both performance management and social support, employees are more likely to behave ambidextrously than in an organization that neglects one part (Birkinshaw & Gibson 2004).

Since Gibson and Birkinshaw (2004) introduced the concept of contextual ambidexterity, several scholars demonstrated the significance of individuals for achieving ambidexterity in an organization (Mom et al. 2007, 2009, Rosing et al. 2011, Good & Michel 2013, Kauppila & Tempelaar 2016). Most of the literature on individual ambidexterity has primarily focused on the managerial perspective (Mom et al. 2007, 2009, Rosing et al. 2011). Mom et al. (2007) underline the pivotal role of managers’ ambidexterity within an organization’s pursuit to become ambidextrous. They research the influences on a manager to either have a more explorative or more exploitative behavior. More precisely, their study focuses on the impact of knowledge inflows on the extent to which managers conduct explorative or exploitative activities. They conclude that top-down knowledge inflows enforce managers to engage in exploitative activities, whereas bottom-up and horizontal knowledge inflows positively relate to the extent managers pursue explorative activities. Another study from the same scholars highlights the effects of formal structural and personal coordination mechanisms on managers’ ambidexterity. Thereby, they intend to get a deeper understanding

(16)

of variations in managers’ ambidextrous behavior. Their results indicate that managers’ ambidexterity is positively influenced by their decision-making authority. Moreover, concerning the personal coordination mechanisms, the scholars claim that both the participation of managers in cross-functional interfaces as well as their connectedness to other organization members positively relate to managers’ ambidextrous behavior (Mom et al. 2009). In another study at the managerial level of analysis, Rogan and Mors (2014) consent with Mom et al. (2009) on the importance of managers’ networks for their ambidextrous behavior. It is an important implication of their paper that the constitution of networks at the individual level can affect organizational level performance. For example, they found evidence that informal ties within a manager’s network increase organizational ambidexterity. On the other hand, formal ties tend to push managers solely to exploitation. Furthermore, and in accordance with the contextual ambidexterity approach, scholars have argued that managers can positively influence employees’ ambidextrous behavior by combining strong managerial support with high performance expectations – also known as paradoxical leadership (Birkinshaw & Gibson 2004, Rosing et al. 2011, Kauppila & Tempelaar 2016). Rosing et al. (2011) concentrate on the leadership-innovation relationship and suggest that the two complementary sets of opening and closing leader behavior foster ambidexterity among individuals and teams. Accordingly, an ambidextrous leader is able to manage the switch between both behavior sets. Hence, ambidexterity at the individual level is necessary in order to cope with the setting of constantly changing requirements of innovation processes.

Besides the focus on behavioral characteristics of managers and leaders, a few papers on individual ambidexterity examined the influence of cognitive abilities (Good & Michel 2013, Kauppila & Tempelaar 2016). In their study, Good and Michel (2013) investigate the impact of different cognitive abilities on task adaptive performance in a dynamic real-time context. The results suggest that particularly the degree of intelligence accounts for significant variance on an individual’s task adaptive performance. Additional variance can be explained

(17)

through variables such as divergent thinking, focused attention and cognitive flexibility. Kauppila and Tempelaar (2016) demonstrate that, besides the leadership style within an organization, psychological factors can predict individuals’ ambidextrous behavior. More specifically, they find out that general self-efficacy enforces individual ambidexterity. This positive relationship is mediated by the learning orientation of an individual.

The limited empirical data and researched areas in the individual ambidexterity literature leave a lot of room for future research in this field. According to Good and Michel (2013), the influence of contextual features on individual ambidexterity represents one of the most crucial areas for further research. As already presented above, the construct of contextual ambidexterity demonstrates that the organizational context can have a huge impact on employees’ ambidextrous behavior (Gibson & Birkinshaw 2004). Good and Michel (2013) particularly emphasize on contextual features in dynamic contexts. They argue that individuals in organizations are confronted with an increasing amount of uncertainty and change. These conditions raise the necessity for individual ambidexterity in dynamic decision-making contexts. Thus, a deeper understanding of the influence of contextual factors on individuals’ ambidextrous behavior will help an organization to better prepare for future challenges. More specifically, Good and Michel (2013) call for future research on the influence of time pressure as a contextual factor on individual ambidexterity.

2.3 Time pressure and individual ambidexterity

Time pressure is a ubiquitous feature for organizations in dynamic contexts. Thus, many employees have to cope with a varying or constantly existing level of time pressure at work – also called a time famine. In this time famine people feel to have not enough time to finish their daily tasks (Amabile et al. 2002). The ubiquity of time pressure has motivated several scholars to research the effects of this contextual feature on work performance, or more specific on explorative behavior. Results of these studies suggest that time pressure can be both positively

(18)

and negatively related to individuals’ explorative work behavior (Kelly & Karau 1993, Simonton 1999, Amabile et al. 2002, Beck & Schmidt 2013, Blank et al. 2014, Hsiao et al. 2015). Kelly and Karau (1993) made an interesting observation when they examined both the initial and persisting effects of time pressure on group creativity. While short initial time limits induced faster rates of performance with a lower level of creativity, persistently high time pressure led to an overall increase in creativity. Hsiao et al. (2015) also found a positive influence of time pressure on creativity. They investigated the relationship between both concepts in industrial design and found a positive relation between short-term time pressure and industrial design creativity. Thus, they even recommend industrial design companies to create appropriate time pressure to increase creativity. Blank et al. (2014) support the positive effects of time pressure on explorative activities by showing that high time pressure can increase radical innovation. They draw the conclusion that managers should not be concerned with negative effects of high time pressure on their team’s radical innovations, but in fact should leverage on the beneficial impact of time limitations for the innovation process. By contrast, Simonton (1999) and Amabile et al. (2002) demonstrate that time pressure can also undermine creative cognitive processes. They point out the importance of incubation time during creative processes. This time is needed to rearrange relevant mental factors into new patterns and subconsciously determine the most likely circumstances among them. Every creative outcome is preceded by distinctive creative cognitive processes. Whenever a person intends to create innovative or creative output, they must identify and prepare the respective problem or gap beforehand. Subsequently, a variety of responses has to be generated and validated. Especially the response generation is both time-consuming and important since typically only a high amount of generated responses enables to derive at a truly creative outcome. As time pressure causes an insufficient amount of time to generate all these responses, people narrow down the information they process in order to reduce complexity at a mental level. This ultimately leads to a deficient creative output. Although Amabile et al. (2002) admit that a certain degree of

(19)

time pressure can enforce people to do more work, they are convinced that high time pressure leads to burnout in the long run while diminishing creativity in the short run. Accordingly, they demonstrate that time pressure, either externally- or self-imposed, creates conditions in which the process of creative thinking simply receives insufficient attention, given all the other demands an individual is facing. Beck and Schmidt (2013) provide another indicator that time pressure can negatively affect explorative work behavior. Their study among undergraduate students identifies perceptions of time pressure as a predictor of state goal orientations. More specifically, their research reveals that high levels of perceived time pressure lead to a decrease of focus on explorative goal orientations, such as developing new skills, as these activities are rather seen as a kind of luxury that should only be completed once the individual does not perceive any time pressure.

When reviewing prior research on the influence of time pressure in organizations, time pressure can be summarized as a crucial contextual factor that directly affects work performance and especially explorative work behavior. Nevertheless, the above presented studies show that the exact impact of time pressure on people’s explorative work behavior is ambiguous and does not allow to draw any one-sided conclusion (Beck & Schmidt 2013, Hsiao et al. 2015).

Even though time pressure research examined the effect on different work behaviors, no study investigated the influence of time pressure on individual ambidexterity so far. None of the above-mentioned studies consider the balance relation between explorative and exploitative activities within changing time pressure scenarios. However, albeit the contradicting results on the relationship between time pressure and explorative work behavior, the studies provide certain implications for the analysis of individual ambidexterity under time pressure. Besides the fact that each research paper draws a different conclusion on the exact influence of time pressure, they all share the same implication: increasing time pressure changes the way people work (Amabile et al. 2002, Beck & Schmidt 2013, Hsiao et al. 2015).

(20)

In accordance with the conceptualization of exploration and exploitation in the first section of this chapter, both concepts are equally important for the survival and success of an organization (March 1991). Consequently, it can be assumed that people will balance explorative and exploitative activities if they do not face any time pressure. By additionally taking into account the fact that exploration and exploitation are two ends of a continuum, it can be presumed that any change to the initially balanced setting without time pressure will change the balance between exploration and exploitation into a less evened ratio. If a person tends to focus more on exploitative activities when facing higher levels of time pressure, then there is less capacity to engage in explorative tasks, and vice versa. In line with Cao et al.’s (2009) balance dimension of ambidexterity, a change of the initially balanced setting without time pressure will thus lead to a lower level of individual ambidexterity as the absolute difference of exploration and exploitation increases. Based on this assumption, the above presented studies about the effect of time pressure on explorative work behavior suggest that increasing time pressure negatively affects the level of individual ambidexterity. This leads to the first hypothesis:

Hypothesis 1a: Time pressure is negatively related to the level of individual ambidexterity.

The literature review on time pressure demonstrates that there are different possibilities how time pressure can affect individuals’ explorative work behavior. Nonetheless, most of the studies do not apply the balance dimension of ambidexterity and therefore do not consider a trade-off between exploration and exploitation. This means that studies like the one from Blank et al. (2014) point out that high levels of time pressure will increase the focus and quality of explorative activities, such as radical innovation, without concluding any consequence for the focus and quality of exploitative activities on the other hand. Beck and Schmidt’s (2013) study comes closest to this trade-off. Their results indicate that individuals will only invest time in

(21)

explorative activities when they do not perceive high time pressure. In accordance with this outcome and the fact that exploration and exploitation are two ends of a continuum, it can be assumed that high levels of perceived time pressure not only decrease the focus on explorative activities but simultaneously increase the focus on exploitative activities. It follows:

Hypothesis 1b: Time pressure is positively related to a focus on exploitative activities.

2.4 The moderating role of resilience

When analyzing the relationship between time pressure and individual ambidexterity, it is important to include individual characteristics as they can explain and predict differences in individuals’ work behavior (Gupta et al. 2006, Good & Michel 2013, Kauppila & Tempelaar 2016). The moderating role of resilience is examined in this study as it represents a personality trait that enables a person to cope with high levels of stress. More precisely, individuals who obtain a high degree of resilience are able to use positive mental skills to handle high pressure and demands and thus remain psychologically steady and focused (Ong et al. 2006, Kinman & Grant 2011, Juster & Marin 2013, Shatté et al. 2017). This ability can be directly linked to a scenario in which individuals face increased stress in form of high time pressure.

Ong et al. (2006) argue that differences in psychological resilience can explain significant variation in responses to daily stress. Their findings indicate that resilience serves to strengthen resistance to daily stress. According to them, high-resilient individuals obtain the capacity to maintain a separation of positive and negative emotional states when facing stress. This allows them to manage complex and stressful situations in a more focused manner. Shatté et al.’s (2017) study supports these observations and adds that high levels of resilience have a beneficial impact on employees’ perception of stress as well as on job-related behaviors related to stress, regardless of a difficult work environment. Accordingly, when facing difficult work environments, high-resilient employees are more productive and better able to avoid absences

(22)

than employees with low resilience. Although current studies on resilience do not form any consensus on what characteristics enable a person to become resilient, several research papers assume that there are certain individual factors that promote resilience (Kinman & Grant 2011, Juster & Marin 2013). Kinman and Grant (2011) ascertain that several emotional and social competencies such as reflective ability and emotional intelligence foster resilience. Juster and Marin’s (2013) paper deals with “protective factors” (p.10) of resilience that can predict psychological adjustment to adversity and improve a person’s response to it. Among the most important factors they name optimism, a proactive coping mechanism and effective emotional regulation.

Summing up the presented findings on resilience, it can be concluded that high-resilient individuals are better able to cope with stressful, demanding situations and remain more focused than individuals with low resilience. Since time pressure represents a concept which increases demands relative to the remaining time, a moderating effect of resilience on the relationship between time pressure and individual ambidexterity can be predicted. Individuals who are able to stay focused when facing high levels of time pressure can be assumed to behave more ambidextrously than individuals who easily stress out in these situations and act irrationally (Ong et al. 2006, Shatté et al. 2017). Thus, the following hypothesis can be deduced:

Hypothesis 2: The relationship between time pressure and individual ambidexterity is negatively moderated by resilience, so that the effect of time pressure on individual ambidexterity is mitigated for high values of resilience.

2.5 The moderating role of openness to experience

Openness to experience is the second personality characteristic that is investigated as potential moderator of the relationship between time pressure and individual ambidexterity. It is part of the Big Five personality traits, also known as the five-factor model developed by Goldberg

(23)

(1990). People who are open to experience can be described as imaginative, broad-minded, curious and perceptive (Barrick & Mount 1991, Barrick et al. 2001). When relating openness to experience to individual ambidexterity, the existing literature provides similar findings for the relationship between both concepts (Lepine et al. 2000, Zacher et al. 2014, Keller & Weibler 2015). Lepine et al. (2000) observed that openness to experience positively shapes individuals’ explorative behavior. More concretely, their study about adaptability to changing task contexts demonstrates that individuals who score high on openness to experience tend to follow new and unconventional ideas. Overall, this personality trait enabled those individuals to make better decisions than those individuals with low openness when facing frequently changing task contexts. Zacher et al.’s (2014) results accord with these findings indicating that openness to experience is the strongest predictor of employees’ exploration behavior within their research project. While mainly investigating the effects of opening and closing leadership on employees’ ambidextrous behavior and innovative performance, they found another interesting effect of openness to experience. Expanding the direct link to explorative behavior, their results show that openness to experience surprisingly also belongs to the strongest predictors of employees’ exploitative behavior. From this perspective, it can be assumed that individuals high on openness to experience are likely to perform ambidextrously. Keller and Weibler (2015) collected supporting data for this assumption. Their research paper identifies openness to experience as negative moderator of the relationship between individual ambidexterity and cognitive strain. Consequently, individuals who are open to experience perceive less cognitive strain when acting ambidextrously. Their strength in divergent thinking and flexibility enables them to better manage the tension between exploration and exploitation. Contradicting these findings, Walsh et al. (2008) did not find any significant effect when testing the moderating role of openness to experience on the relationship between time pressure and creativity. However, their investigation only focused on explorative behavior whereas this study examines individual ambidexterity.

(24)

Thus, putting the above stated findings on openness to experience into context with the main relationship between time pressure and individual ambidexterity, it can be presumed that individuals who score high on openness will behave more ambidextrously under time pressure than individuals who have low values in this characteristic. This leads to the last hypothesis:

Hypothesis 3: The relationship between time pressure and individual ambidexterity is negatively moderated by openness to experience, so that the effect of time pressure on individual ambidexterity is mitigated for high values of openness to experience.

2.6 Theoretical framework

Including all reviewed literature and the deduced hypotheses, a conceptual model can be drawn (figure 1). The model visualizes the relations between the different variables and gives an overview of the conducted research.

(25)

3. Data and method

This chapter presents the methodology and statistical framework of the conducted vignette study. This includes the reasons for choosing the vignette format for the data collection, statistical insights about the participant group and how the investigated variables of this research are measured. Lastly, the statistical framework of the study is assessed which provides the basis for the results section in the following chapter.

3.1 Sampling strategy

In order to quantify and test the above stated hypotheses, data were collected via a quantitative vignette study (see appendix 1). A vignette study consists of two building blocks: (i) a vignette experiment as core component, and (ii) a traditional survey format to gain supplementary information on additional participant-specific characteristics. Subsequently, these additional information can be used as covariates to the core data generated in the experiment (Atzmüller & Steiner 2010). A vignette experiment presents hypothetical scenarios to which participants of the study respond. The vignette uses short descriptions of situations typically containing tasks which the participants need to solve. Thereby, it reveals participants’ perceptions of events and elicit their judgments about these scenarios (Hughes & Huby 2004, Atzmüller & Steiner 2010). The format of the vignette study was chosen because first of all, it ensures data collection in a dynamic setting as participants are confronted with a dynamic real-life scenario. Thus, the study is able to respond to calls from the ambidexterity research field to examine the construct of individual ambidexterity in a dynamic context (Good & Michel 2013). Secondly, a vignette study is, similar to the traditional survey format, a good method to easily collect a large amount of quantitative data from a widely distributed population in a highly economical way. Furthermore, it facilitates to standardize data in a simple way allowing for easy comparison among participants (Saunders et al. 2012).

(26)

For optimal generalizability of the study results, it was necessary to achieve a heterogeneity as high as possible among participants. All analyzed concepts and variables of the research project can be tested among individuals with different backgrounds since they do not require any specific knowledge. Consequently, it was possible to reach a greatly heterogeneous group of respondents.

In order to prevent potential misunderstandings or flaws in the vignette study and to additionally pre-check for validity of the presented scales, the vignette was pilot tested (see appendix 2). During the pilot phase 20 individuals answered the vignette study. The outcome of the pilot test was both positive and helpful. All necessary scale validity was reached and additionally some respondents gave constructive feedback on improving the phrasing of certain items to avoid misinterpretations. The pilot results also indicated that the completion of the vignette study takes approximately eight to ten minutes, and at least four minutes to properly read all questions and scenario instructions.

3.2 Sample

The data collection of the vignette study was conducted online and hence each participant received an email with a link that guided the respondent directly to the vignette test. In total, the vignette study was sent out to 201 individuals of which 151 completed the entire test. During the process of data cleaning, all variables of the study were checked in terms of missing values and time spent for completing the experiment. Since the vignette study was programmed in such way that each question within the vignette was followed by a mandatory answer, the performed frequency check did not reveal any missing values among the completed exemplars. However, some participants finished the vignette in a lapse of time that was considered as not enough to have properly read all instructions and answered all questions – they completed it in

less than four minutes. Therefore, the total amount of 151 observations was reduced to 135

(27)

The 135 respondents were composed of an almost equal amount of male and female individuals – 49.6% male, 50.4% female. The average age of the participants was 31.33 years with a standard deviation (SD) of 12.74 and their average work experience amounted to 6.89 years (SD = 10.01). In terms of occupation, students represented the largest group of participants (68), followed by individuals with an entry level position (23) and associates (23). Among the respondents who currently work, most are employed in the education sector (19), closely followed by the automotive industry (18) and the consultancy business (17).

In order to reveal all potential factors influencing the final results, these statistics should be looked at from a divided perspective – divided in the control group and the experimental group. From the total 135 respondents, 69 individuals were part of the control group and thus completed the first scenario without time pressure. These 69 respondents were composed of 60.8% female and 39.2% male individuals. The average age of this group was 29.11 years (SD = 13.01) with an average work experience of 5.69 years (SD = 9.78). Besides a large number of students (39), most of the individuals of the control group are working as an intern (11) or at the entry level (10). Among the already employed respondents of the group, most of them work in the education industry (14), followed by the finance industry (9) and the retail sector (6).

The remaining 66 individuals constituted the experimental group and thus faced the second scenario with increased time pressure. Contrary to the control group, the 66 respondents of the experimental group were composed of more men (59.3%) than women (40.7%). Additionally, they were, on average, older than the individuals of the control group (33.64 years, SD = 12.21) and already gained more work experience (8.14 years, SD =10.32). The predominant number of individuals in the experimental group are also students (29), followed by persons working as associates (14) and at an entry level position (13). Unlike the industry distribution of the control group, the respondents of the experimental group predominantly work in the automotive industry (14), followed by the consultancy business (10) and the health services sector (5).

(28)

3.3 Measures

3.3.1 Independent variable

The independent variable of this study is time pressure. Together with the dependent variable individual ambidexterity, the construct of time pressure was created and applied via the self-constructed vignette experiment. For the experiment, two different scenarios were created – one with hypothetical time pressure, one without. In both scenarios participants were facing a situation in which they were, in the role of a firm’s product manager, fully responsible for the success of a certain product within the company they were working at. After having read the complete scenario description, each participant was instructed to decide on the order in which he or she will execute the remaining product management tasks – the focal tasks will be highlighted in more detail in the next section concerning the dependent variable. The only point in which both scenarios differed was the level of time pressure. In the first scenario, participants were provided with the information that they will have enough time to complete all given tasks. Consequently, no time pressure was present for the execution of the tasks. In contrast, the second scenario indicated that the CEO of the company would critically check the short-term profitability and long-term potential success of the focal product within the next days. As each of the remaining 14 tasks took one full day to be completed, it was certain that the participant is not able to finish all tasks before the visit of the CEO. Thereby, time pressure was created. This method of creating hypothetical time pressure was adopted from a former study that has successfully manipulated participants that way (DeDonno & Demaree 2008).

The participants were divided into two groups – a control group and an experimental group. The participants were randomly assigned to each group. The difference between both groups is that the experimental group is facing some form of planned manipulation while the control group is not manipulated (Saunders et al. 2012). Accordingly, the control group was assigned to the first scenario without time pressure while the experimental group worked on the second scenario in which time pressure was manipulated. Thus, a between-subject design was

(29)

present in the vignette experiment whereby each participant went through only one experimental treatment. For optimal comparability and measurement of the effect of time pressure, it was important to create two identical scenarios that only differ in terms of time pressure. By doing so and randomly assigning the participants to the control and experimental group, it was possible to control and remove possible influence of an alternative explanation to the manipulated time pressure. This erases threats to internal validity (Saunders et al. 2012). To additionally foster internal validity of the independent variable, a manipulation check was included in the vignette study. Such a check is a common tool within vignette experiments to verify the successful manipulation of the respective variable (Ordoñez & Benson 1997, Maule et al. 2000, Crescenzi et al. 2013). The manipulation of time pressure was checked by asking the participants, upon completion of the experiment, if the uncertainty about the deadline of their task influenced their choice. The question was answered on a four-point Likert scale ranging from “Not at all” (= 1) to “To a great extent” (= 4). The mean of 3.18 shows that time pressure was successfully manipulated, in the sense that the respondents belonging to the experimental group were significantly influenced by the created time pressure in the second scenario.

3.3.2 Dependent variable

As already mentioned above, individual ambidexterity represents the dependent variable within the conceptual model of the study. To measure the ambidextrous behavior of each participant, 14 tasks were created – seven explorative, seven exploitative – which the participant had to rank in the vignette experiment. Both scenarios provided incentives to care about the exact order of the tasks. In scenario one, participants were advised that, depending on their choice to rank, certain tasks will be completed later than the others which might negatively affect the success of their outcome. In the second scenario, the participants had to consider their choice very carefully as they did not know how many tasks they were able to complete in the remaining

(30)

time before the visit of the CEO. For example, if the CEO decided to already make a visit after five working days, the participant (hypothetically) had to convince the CEO of the focal product on the basis of the five highest ranked tasks.

To ensure internal scale validity, all content for the scenario tasks was based on Mom et al.’s (2009) validated 14-items scale for individual ambidexterity. In terms of internal validity, the seven explorative items indicate a Cronbach’s alpha of 0.90, the seven exploitative items one of 0.87. Thus, the scales obtain a high internal consistency. In order to tailor these items to the specific tasks of the scenarios, each item content was adjusted accordingly. Providing an example for the undertaken adjustments, Mom et al.’s (2009) item “Evaluating diverse options with respect to products/services, processes or markets” was replaced by “Evaluate diverse innovative techniques to produce the product in a cheaper way”. In order to further check for validity of the 14 items, an additional validity check was performed during the above-mentioned pilot phase of the vignette study. The validity check provided the respondents with a definition of both exploration and exploitation and asked them to sort each of the 14 items to one of the concepts based on the revealed information. With this method, it was possible to check whether the respondents share the same understanding of the concepts of exploration and exploitation. The result of the validity check was helpful as it showed that the respondents predominantly shared the same understanding of the dependent variable. On average, they sorted 11.8 of the 14 items correctly. Based on the participants’ feedback, an analysis and adjustment of the items, which were most often sorted in the wrong category, was feasible. Thereby, potential flaws and misunderstandings of the concepts could be prevented for the final vignette experiment.

The exact level of individual ambidexterity was calculated by assigning each task rank a specific amount of points, such that the first ranked task of a participant was worth 14 points, the second one 13 points, the third one 12 points, and so on. After completion of the experiment, all points for explorative and exploitative tasks were summed up separately. The quotient of

(31)

the exploration sum divided by the exploitation sum finally indicated the level of individual ambidexterity. In accordance with the balance dimension of ambidexterity, the individual ambidexterity of participants reaches high levels when the absolute difference of exploration and exploitation is low (He & Wong 2004, Cao et al. 2009). Thus, if the quotient is close to or equal to one (» 1), the respective participant behaved remarkably ambidextrous because he/she managed to balance explorative and exploitative tasks the best possible way. For any result below one (< 1), the participant focused more on exploitative tasks than on explorative ones, and vice versa. A sample calculation is provided in appendix 3.

3.3.3 Moderating variables

The personality traits resilience and openness to experience embody the two potential moderating factors in this study. The level of participants’ resilience was measured on the 10-items Connor-Davidson Resilience Scale (Aloba et al. 2016). The respondents answered the ten questions on a five-point Likert scale, ranging from “Never” (= 1) to “Always” (= 5). A sample item of the construct is: “Under pressure I stay focused”. The scale does not contain any counter-indicative items. Consequently, a relatively high average score on the scale represents high levels of a participant’s resilience. The reliability of the scale was checked by calculating the Cronbach’s alpha. The result of α= 0.782 indicates a high level of internal consistency.

Additionally, the corrected item-total correlation for all items of the variable were above 0.30. Therefore, no item had to be deleted.

The openness to experience of participants was gauged on a validated IPIP-NEO scale including four different items (Goldberg 1999, Johnson 2014). Alike the resilience scale, the items were answered on a five-point Likert scale, ranging from “Never” (= 1) to “Always” (= 5). A sample item of the construct is: “Prefer variety to routine”. The scale comprised counter-indicative items, such that an agreement with those items represents a low level of the construct. These counter-indicative items were recoded during the data analysis. Accordingly, a relatively

(32)

high average score on the four items represents high levels of a participant’s openness to experience. The scale had a sufficient level of internal consistency as the Cronbach’s alpha equaled α= 0.713. As well as the resilience items, the corrected item-total correlation for all

items of the variable were above 0.30. Therefore, all items could be applied.

3.3.4 Control variables

The control variables that were used in this study are: age, gender, work experience, occupation and the industry the participant currently works in. Control variables were included in the vignette study to ensure that all results for the dependent variable were caused by the above-mentioned independent variable and moderating variables and not influenced by any other factor. Furthermore, it was important to receive detailed information about the respondents to assess the heterogeneity of the sample (Saunders et al. 2012).

All control variables were assumed to be potential influencers for the investigated relationship between time pressure and individual ambidexterity. This assumption derived from the results of several research papers on organizational context and ambidextrous behavior (Mom et al. 2015, Sturman 2003, Brion et al 2010, Gibson & Birkinshaw 2004, Ghoshal & Bartlett 1994). Besides the relevance of age and gender, Mom et al. (2015) and Sturman (2003) especially point out the significant influence of people’s work experience on their (ambidextrous) work behavior. Within the field of organizational context embedded in the ambidexterity research area, the variables occupation and industry were identified as determining factors for ambidextrous work behavior (Brion et al 2010, Gibson & Birkinshaw 2004, Ghoshal & Bartlett 1994).

3.4 Statistical framework

After data collection via the online vignette study, the final data were checked and prepared for the statistical analysis. Besides the already mentioned process of data cleaning, recoding of

(33)

counter-indicative items and scale reliability checks, this included testing for normality, skewness and kurtosis of the collected data. The normality test of the variables indicated a normal distribution for all, with skewness and kurtosis values within the common rule of thumb (between -0.5 and 0.5).

Moreover, a factor analysis was carried out for all respective variable scales. By means of a factor analysis, the shared variance between variables can be examined, based on the assumption that the variables are determined by the presence of latent not-observable dimensions. For all investigated items the Kaiser-Meyer-Olkin measure verified the sampling adequacy for the analysis, as KMO = 0.819. Furthermore, Bartlett’s test for sphericity indicated that the correlations between all items were sufficiently large for executing the factor analysis (χ2 (91) = 431.152; p < 0.001). According with Kaiser’s criterion, the factor analysis found three components that had an eigenvalue above 1. In combination, these factors explained 49.29% of the total variance. Nevertheless, all scales were retained in their original form. The results of the reliability analysis support the retention of the original scales since it revealed that extracting factors did not lead to an improved Cronbach’s Alpha for any of the scales.

(34)

4. Results

Building up on the previous chapter, the following sections show the results of the conducted vignette study and the underlying data analysis executed in SPSS. Firstly, descriptive statistics for the variables of the study are provided. Additionally, a correlation analysis is conducted to show how the variables are related to each other. In the second part, a one-way analysis of variance (ANOVA) is performed. This analysis allows to test the first two hypotheses which are related to the differences of means between the control group and experimental group of the vignette study. Subsequently, a hierarchical regression analysis is done to reveal the direct effect of time pressure on individual ambidexterity after excluding the influence of the control variables. Finally, a moderation analysis is conducted to test if the hypotheses concerning the moderating role of resilience and openness to experience hold true.

4.1 Descriptive statistics and correlation analysis

Table 1 presents descriptive statistics and the bivariate correlation analysis of all investigated variables. The descriptive statistics comprise the mean and standard deviation (SD) for each variable of the study. In order to calculate those numbers, scale means were computed for all items that were used to describe a variable. Deriving from the mean of each variable, it was possible to assess the level of correlation between the variables. The relationships were investigated using the Pearson correlation coefficient (r). In the table, all significant correlations are marked with one or two asterisks (*/**) depending on the level of significance (p).

The correlation analysis indicates a positive and significant correlation between the moderating variables resilience and openness to experience, with a Pearson correlation coefficient of r = 0.47 at a significance level of p < 0.01. Furthermore, the table shows that the independent variable time pressure is positively and significantly correlated with the control variable age (r = 0.18; p < 0.05). Additionally, and even more relevant, the table reveals that time pressure is negatively and significantly correlated with the dependent variable individual

(35)

ambidexterity (r = -0.36; p < 0.01). Between time pressure and the two moderating variables as well as the other control variables no significant correlation coefficient is present. The same holds for individual ambidexterity. Among the control variables the table provides further details concerning the sample configuration. Besides the obvious fact that age is positively and significantly correlated with work experience (r = 0.93; p < 0.01), the correlation coefficient between gender and work experience is also positive and significant (r = 0.19; p < 0.05). In line with these findings, age and gender also reveal a positive and significant correlation (r = 0.26;

p < 0.01). Thus, male participants were on average older and had more work experience than

female respondents participating in the study. The correlation numbers between the above-mentioned control variables and occupation and industry are not worth interpreting. The variables occupation and industry themselves were valuable for the descriptive analysis of the sample in the previous chapter. Nevertheless, their correlation numbers do not deliver any sense-making results due to the fact that both variables contained multiple possible answers without having a linked meaning between the content of the possible answers and the respective number assigned to each answer option.

(36)

Table 1: Means, Standard Deviations, Correlations Variables Mean SD 1 2 3 4 5 6 7 8 9 1. Age 31.33 12.74 - 2. Gender 1.50 0.50 0.26** - 3. Work Experience 6.89 10.01 0.93** 0.19* - 4. Occupation 3.12 2.49 0.77** 0.19* 0.78** - 5. Industry 8.24 4.70 -0.25** -0.09 -0.27** -0.40** - 6. Resilience 3.72 0.49 -0.05 0.18* -0.03 -0.03 -0.09 - 7. Openness to Experience 2.62 0.55 -0.03 0.10 0.02 0.08 -0.11 0.47** - 8. Time Pressure 0.40 0.49 0.18* 0.16 0.12 0.12 -0.05 0.08 0.13 - 9. Individual Ambidexterity 0.87 0.34 0.05 0.04 0.08 -0.05 -0.01 0.02 0.11 -0.36** - N = 135

**. Correlation is significant at the 0.01 level (2-tailed). *. Correlation is significant at the 0.05 level (2-tailed).

The effect of time pressure on individual ambidexterity : a vignette study

Master Thesis