Openness to experience and individual ambidexterity : the moderating role of context

Openness to experience and individual ambidexterity:

The moderating role of context

Mark de Weers

10694382

MSc Business Administration – Strategy track

Supervisor: Pepijn van Neerijnen

i Statement of Originality

This document is written by Mark de Weers 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.

Utrecht, 29 June 2015 Mark de Weers

ii

Table of contents

I. Introduction 1

II. Theory and hypotheses 4

III. Method 14 IV. Results 21 V. Discussion 26 VI. Conclusions 33 VII. References 35 VIII. Appendix 44

1

Openness to experience and individual ambidexterity:

The moderating role of context

Mark de Weers

MSc Business Administration – Strategy track

By taking a trait perspective this research attempts to extend the current body of knowledge on individual ambidexterity. This research examined the role of openness to experience in relation with individual ambidexterity. In addition, this research also examined the moderating role of context. The results of this study are based on data collected from 455 individuals who played a business simulation game in teams of 3 to 5 students. As confirmed by the hypothesis, individuals who are more open to new experiences are better able to act ambidextrous. This study also shows that role ambiguity has a positive significant moderating effect on the relationship between openness to experience and individual ambidexterity. Need for cognition, creative problem solving and reciprocal task interdependence did not have a significant moderating effect on the main relationship. Surprisingly, reciprocal task interdependence did have a positive direct effect on individual ambidexterity.

Keywords: Individual ambidexterity, openness to experience, role ambiguity, need for cognition, creative problem solving, reciprocal task interdependence.

I. Introduction

It is important for a firm to exploit its current knowledge while at the same time explore possibilities for the future, to keep aligned with the ever changing environment (March, 1991). This phenomenon called organizational ambidexterity has been an important research topic (Gibson & Birkinshaw, 2004; Raisch, Birkinshaw, Probst, & Tushman, 2009). A reason

2 for this might be that organizational ambidexterity is associated with sustained firm performance (Keller & Weibler, 2014; Raisch et al., 2009; Smith & Tushman, 2005). The way this can be achieved depends according to Laureiro-Martinez, Brusoni, and Zollo (2010) next to organizational arrangements and collective processes on the actual capacities and behaviours of individual members of the organization. Confirmed by Mom, van den Bosch, and Volberda, (2009) who stated that in order for a firm to become ambidextrous one needs ambidextrous managers, thereby indicating the importance of the individual ambidexterity. According to Mom, Fourné and Jansen (2015) ambidexterity causes internal tensions because exploration and exploitation require to think in terms of contradictions. This ostensible contradiction in the concept of ambidexterity can be described as a paradox (Smith & Lewis, 2011). Although research has already shown that ambidexterity leads to higher performance of a firm, business unit or an individual (O’Reilly & Tushman, 2013), it is still unclear why some individuals are better able to deal with the paradox of ambidexterity than others. Therefore, it is important to develop a greater understanding of individual ambidexterity.

The exploring and exploiting behaviour of individual members of an organization is a growing research topic. Mom, Van Den Bosch, and Volberda (2007) have taken an important step in researching individual ambidexterity by investigating how the acquisition of knowledge by managers influences exploration and exploitation activities. Although their findings are valuable for the individual ambidexterity literature, it does not provide a greater understanding why some individuals are better able to act ambidextrous. This is supported by Bonesso, Gerli, and Scapolan (2014) who state one of the limitations by (Mom et al., 2009) is that they neglected personality characteristics in their research on individual ambidexterity. In order to answer why some people engage more in exploration and exploitation activities it is required to explore employees’ personal characteristics (Gupta, Smith, & Shalley, 2006; Raisch et al., 2009). This is particularly important since exploration and exploitation activities

3 are based on actor’s decisions, and decisions of individuals are influenced by personality characteristics (Lepine, Colquitt, & Erez, 2000). A personality characteristic which might be especially important for the individual ambidexterity research is openness to experience. Namely Keller and Weibler (2014) found that persons who score high on openness to experience perceive less cognitive strain from acting ambidextrously, because they think divergent (exploration vs. exploitation). The openness to experience dimension consists of persons who are intellectual, cultured, curious, imaginative, original, broad-minded, intelligent, and artistically sensitive (Barrick & Mount, 1991; Barrick, Mount, & Judge, 2001). Thereby they would be better able to handle the tension between exploration and exploitation. Therefore, this research argues that openness to experience is positively related to individual ambidexterity.

Another aspect which needs to be considered is the context in which the relationship is measured. This is often neglected by research in the past (Griffin, 2007). Thus, the context in which the relationship between openness to experience and individual ambidexterity is measured. This is supported by Good and Michel (2013) who state that individual ambidexterity should be considered in a context specific way. This research aims to include context by adding role ambiguity, need for cognition, creative problem solving and reciprocal task interdependence as moderating variables. Thereby responding to for example Jundt, Shoss, and Huang (2014, p. 6) who state that “ Instead of focusing on Big Five traits’ main effects, researchers need to offer a clear understanding of person–situation interactions by involving contextual and job level moderating variables”.

The goal of this study is to continue and deepen research on individual ambidexterity since research within this area is far from complete (Lavie, Stettner, & Tushman, 2010; Raisch et al., 2009). By taking a trait approach this research contributes to the current literature on individual ambidexterity. As research falls short in explaining why some

4 individuals are better able to behave ambidextrous than others, it is important to continue the research on individual ambidexterity. As suggested personality characteristics may play a key role, and particularly openness to experience might be the reason why some individuals are better able to act ambidextrous.

Secondly, this research contributes to the current body of knowledge on individual ambidexterity by responding to the call to examine the role of context (Lavie et al., 2010; Raisch et al., 2009). This study suggests that the ability to act ambidextrous is dependent on different parts of an individual’s work context such as: role ambiguity, need for cognition, creative problem solving and reciprocal task interdependence. In the following section the theory and hypothesis that form the basis of this study are presented.

II. Theory and hypotheses

Individual ambidexterity can be described as the behavioural actions from employees by exploring long-term organizational systems, experimenting, innovating and reconsidering existing beliefs and decisions, while at the same time exploiting their existing knowledge, applying, improving, and extending existing competences and elaborating on existing beliefs and decisions (Mom et al., 2007). Ambidexterity is made up by two distinct but interrelated concepts. The core of the first concept, exploration, consists of activities related to creating variation in experience which is associated with broadening a person’s existing knowledge foundation (Mom et al., 2007). For instance, pursuing and obtaining new knowledge and other activities like risk taking, experimentation, playing, being flexible, discovering and innovation search (Mom et al., 2007; Raisch & Birkinshaw, 2008). Exploitation on the other hand refers to the improvement of the reliability of experiences and the integration and deepening of a person’s knowledge (Keller & Weibler, 2014; Mom et al., 2007). Exploitation has to do with

5 activities such as refinement, efficiency, selection, implementation, production and execution (Mom et al., 2007; Raisch & Birkinshaw, 2008).

At least two distinctions in the ostensible contradiction in the concept of ambidexterity can be made (Smith & Lewis, 2011). Smith and Lewis (2011) distinguish among others paradoxes and dilemmas as organizational and individual tensions. A dilemma involves a choice between A and B which is placed on a scale, thereby weighing the pros and cons. A paradox on the other hand consists of contradictory yet interrelated parts (A and B) that exist at the same time and persist over time (Lewis, 2000; Smith & Lewis, 2011). For instance exploration and exploitation seem logical when treated alone but the opposite when juxtaposed. According to Smith and Tushman (2005) a paradox arises when there are tensions in a situation (explore/exploit) and these are juxtaposed through one’s cognition. Thinking in terms of a paradox offers an opportunity to deal with contradictory yet interrelated parts that exist simultaneously and persist over time (Lewis, 2000; Smith & Lewis, 2011). In other words, a person would not have to choose between exploring the new and exploiting the current, but should be able to pursue both.

Most of the literature has focussed on the firm level, unit level or team level, which leaves only modest amount of literature available on individual ambidexterity (Turner, Swart, & Maylor, 2013). A reason for this may be that ambidextrous behaviour at the individual level is difficult to observe (Rogan & Mors, 2014). One of the investigations of ambidexterity on the individual level has been done by Mom et al. (2009). They focussed on formal structural mechanisms and personal coordination mechanisms. They found that a manager’s decision making authority had a positive effect on this manager’s ambidexterity. Subsequently, they showed that manager’s authority and formalization of tasks, and the extent to which managers were connected to other organizational members had a positive effect on ambidexterity. Rogan and Mors (2014) found that a manager’s network is a determinant in being able to act

6 ambidextrously. Good and Michel (2013) used several abilities namely divergent thinking, focussed attention and cognitive flexibility to explain individual ambidexterity. Although the body of research on individual ambidexterity is growing, there are still unanswered questions and uncertainties. The first way of closing this gap, is by taking a trait approach in explaining individual ambidexterity.

Openness to experience and individual ambidexterity

The modest amount of literature available on individual ambidexterity has mostly neglected personality characteristics in explaining why some people engage more in exploring and exploiting activities. This research is particularly interested in the personality trait openness to experience. There are many personality traits affecting the way people experience their lives (DeNeve & Cooper, 1998; Goldberg, 1990). Some of these personality characteristics are well known, such as the big five personality traits (Five Factor Model) (Goldberg, 1993). Persons who are open to new experiences can be described as intellectual, cultured, curious, imaginative, original, broad-minded, intelligent, and artistically sensitive (Barrick & Mount, 1991; Barrick et al., 2001). The first reason why it is meaningful to include openness to experience when researching individual ambidexterity can be derived from the research by Keller and Weibler (2014). They found that transformational leadership and cognitive constrain were positively related to individual ambidexterity. Even more interesting, they found that openness to experience negatively moderated the relationship between individual ambidexterity and cognitive strain. This could mean that people who score higher on openness to experience are better able to deal with the tension of exploration and exploitation. The notion that persons who score high on openness to experience may perceive less cognitive strain from acting ambidextrous because they think divergent is the first indicator that openness to experience is important in explaining individual ambidexterity.

7 As stated earlier, persons who score high on openness to experience are among others intellectual, curious, imaginative, original, broad-minded and intelligent (Barrick & Mount, 1991; Barrick et al., 2001). These characteristics could enable them to act ambidextrous by rethinking current activities and replacing them with alternatives (Keller & Weibler, 2014). By being able to rethink these current activities and replace them with alternatives, persons who score high on openness to experience might be more open to the conflicting demands of exploration and exploitation (Keller & Weibler, 2014). This is a second indicator that openness to experience is important in explaining individual ambidexterity. Persons who are open to new experiences may be able to deal with the paradox of search and refinement.

The research of Keller and Weibler (2014) is a doorway for further research of personality characteristics and individual ambidexterity. In 2006 Gupta et al. (2006) already suggested that individual differences may impact individual ambidexterity. This claim is further supported by Bonesso, Gerli, and Scapolan (2014) who state that one of the limitations by Mom et al. (2009) is the assertion of their contribution on the individual ambidexterity literature while neglecting the personal characteristics of employees which lead to ambidextrous behaviour. According to Lepine et al. (2000) individuals high on openness to experience are curious and search often for new experiences and ideas. In a study by Zacher, Robinson and Rosing (2014), the opening and closing behaviour by leaders was used to predict exploration and exploitation behaviours. Subsequently, exploration and exploitation behaviours were assumed to influence innovative performance. The study controlled for openness to experience as a personality characteristic, since it was expected that persons who score high on openness to experience would have a tendency to engage in more exploration activities. The study by Zacher et al. (2014) found, as expected, that openness to experience was the strongest predictor of employee exploration behaviour. More surprisingly was the fact that openness to experience emerged next to conscientiousness as the strongest predictor of

8 employee exploitation behaviour. What could mean that employees from this study would be able to engage in exploration and exploitation behaviour simultaneously. The study by Zacher et al. (2014) is another indication that persons who score high on openness to experience are better able to act ambidextrous. This leads to the first hypothesis of this study:

Hypothesis 1: Openness to experience is positively related to individual ambidexterity.

Importance of contextual specificity examining the relationship between openness to experience and individual ambidexterity

According to Jundt et al. (2014) researchers should not only focus on the main effects of personality traits, a clearer understanding of person-situation interactions should be provided by including contextual moderating variables. Since individual’s choices do not stand on their own, in a vacuum, they depend on the context in which they operate, thus on the situational or contextual factors (Morgeson & Hofmann, 1999).

According to Borman and Brush (1993) the highest percentage of total common variance accounted for when looking at managerial performance is planning and organizing: formulating short- and long-term goals. Since individual ambidexterity consists of exploitation, achieving short term goals, and exploration, accomplishing long term goals, it is arguable that ambidexterity is a precursor for managerial performance. Tett and Christiansen (2007) stated that situational specificity is an important characteristic when investigating the connection between personality and managerial performance, and thus individual ambidexterity. In addition, Tett and Guterman (2000) state that in order for a trait to be expressed, a situation must occur in which the proposed trait will be activated. In order words, the utterance of a personality trait demands a situation or context relevant to the proposed trait (Tett & Burnett, 2003; Tett & Guterman, 2000).

9 In the interactionist model of Tett and Burnett (2003), the concept of trait activation proposes that the expression of a personality trait depends on the situation. Tett and Burnett (2003) took an important step in researching this phenomenon by providing situational specificity of personality-job performance relations. Although the dependent variable in their interactionist model is job performance and not individual ambidexterity their proposed model is still very valuable for individual ambidexterity research. Since irrespectively of the outcome variable the expression of a trait will still be dependent on the situation. Moreover, Tett and Burnett (2003) stated that job performance among others consists of knowledge, skills and motivation. Mom et al. (2009) point out that one of the characteristics of ambidextrous managers is that they refine and renew their knowledge and skills. Subsequently, Mom et al. (2009) state that ambidextrous managers have the motivation and ability to understand and pursue contradictions. Which is another indication that ambidexterity is a precursor for individual performance. Therefore it is arguable that the personality-job performance relation can be translated into a personality-ambidexterity relation. In the proposed model by Tett and Burnett (2003) several cues: task, social and organizational, moderate the relationship between personality and job performance. These contextual variables can be directly related to the tasks of an employee, the social environment or the whole environment as stated by Tett and Burnett (2003). The model of Tett and Burnett (2003) provides a foundation to hypothesize for several contextual moderating variables.

The moderating role of role ambiguity

Within the model of Tett and Burnett (2003), role ambiguity is on the social level of trait relevant cues. Although role ambiguity does not automatically presumes working with others, without other employees there would be no role ambiguity. Rizzo, House, and

10 Lirtzman (1970, p.156) defined role ambiguity as “reflect certainty about duties, authority, allocation of time, and relationships with others; the clarity or existence of guides, directives, policies; and the ability to predict sanctions as outcomes of behavior”. Rizzo et al. (1970) argued for a decrease of role ambiguity within an organization in order for subordinates to be satisfied and perform well. A framed set of tasks, duties and responsibilities of an employee hold them accountable for their results and performance. Subsequently the tasks, duties and responsibilities would provide the employees with support and guidance. Rizzo et al. (1970) stated that if an employee does not have clear tasks, duties and responsibilities he will be hesitated to make decisions, avoid stressful situations, will be dissatisfied and suffer from anxiety, which will lead to a decrease in performance. But does the same hold true for ambidextrous employees? Or do ambidextrous employees perform better under uncertain circumstances? Mom et al. (2015) stated that ambidextrous employees perform better under increased levels of uncertainty. Since ambidextrous employees are able to combine a long- and short-term orientation, using internal and external knowledge, they will make legitimate choices when they encounter uncertain, unclear and novel problems and situations (Mom et al., 2015). Subsequently, these new experiences lead to the gathering of new information and objectives. In addition, Gelatt (1989) stated that unambiguous goals will not lead to new experiences. So when an employee his role is ambiguous and objectives are not dictated, one can explore new experiences. New experiences such as cognitive challenges, assessment of unexplored solutions and other applications of gathered information may nurture ambidextrous behaviour. This leads to the second hypothesis of this study:

Hypothesis 2: The positive relationship between openness to experience and individual

ambidexterity is moderated by role ambiguity, so that this relationship is stronger for higher values of role ambiguity.

11

The moderating role of need for cognition

Within the model of Tett and Burnett (2003), need for cognition is on a task level of trait relevant cues because need for cognition is applicable to the daily tasks and activities of an employee. There are multiple reasons to expect need for cognition to have a positive moderating effect on the openness to experience-individual ambidexterity relation. The first reason is that individuals high in need for cognition have a tendency to be active, have exploring minds, engaging in cognitive endeavours and due to their intellect get more from their surroundings (Cacioppo, Petty, & Chuan Feng Kao, 1984; Cacioppo, Petty, Feinstein, & Jarvis, 1996). According to Wu, Parker, and Jong (2014) people high in need for cognition have a tendency to engage in (and like) situations which are novel, complex and uncertain and they are effective in linking new and existing knowledge. This implies that people who score high on need for cognition like to explore new possibilities, while at the same time they are highly effective in exploiting due to their ability to link new and existing knowledge.

The second reason is that according to Briñol and Petty (2005) persons with a high score on need for cognition have a tendency to devote all attention to their current task and make use of all information based on empirical and rational deliberation. It is expected that they are more confident about their own thoughts and ideas and that would be helpful championing problems (Wu et al., 2014).

The last reason concerns the relation between need for cognition and openness to experience. Sadowski and Cogburn (1997) showed that there is a positive significant link between need for cognition and openness to experience. According to Sadowski and Cogburn (1997) this relationship reflects the delight of cognitive activity. This would mean that higher values for need for cognition would positively moderate the relationship between openness to experience and individual ambidexterity. Accordingly, it is expected that:

12

Hypothesis 3: The positive relationship between openness to experience and individual

ambidexterity is moderated by need for cognition, so that this relationship is stronger for higher values of need for cognition.

The moderating role of creative problem solving

Creative problem solving is on a task level of trait relevant cues since it encompasses day-to-day activities (Tett & Burnett, 2003). According to Mom et al. (2015) ambidextrous managers need to be able to come up with creative solutions to atone the seemingly conflicting demands of exploration and exploitation. Activities associated with creative problem solving are identification of a problem, obtaining the necessary information, the generation of solutions and the assessment of the generated ideas (Reiter-Palmon & Illies, 2004). This means that people who score high on creative problem solving are better able to identify problems and solve them than people who score lower on creative problem solving. Creative problem solving is according to Gilson and Shalley (2004) depended on the ability of someone to use ideas out of one’s field of expertise and combine multiple sources to examine new alternatives. There are at least two reasons why it is expected that creative problem solving has a moderating effect on the openness to experience-individual ambidexterity relationship. The first reason is because in order to successfully act ambidextrous it is important to be able to solve paradoxical dilemmas related to seemingly conflicting demands in terms of time, risk, strategy and responsibilities (Mom et al., 2015). One way of solving these paradoxical dilemmas is moving along the activities associated with creative problem solving. Therefore people who score high on creative problem solving should be able to be better deal with these paradoxical tensions. The second reason to expect creative problem solving to have a moderating role is because openness to experience is related to creativity (George & Zhou, 2001; McCrae, 1987; Taggar, 2002). According to Wolfradt and Pretz

13 (2001) creative people seem to be more open to new experiences. Therefore, it is hypothesized that:

Hypothesis 4: The positive relationship between openness to experience and individual

ambidexterity is moderated by creative problem solving, so that this relationship is stronger for higher values of creative problem solving.

The moderating role of reciprocal task interdependence

Reciprocal task interdependence is on the social level of trait-relevant cues since it includes the interdependence and interactions between employees (Tett & Burnett, 2003). Reciprocal task interdependence can be described as the need of knowledge provision, making data available and backing from other employees in the organization in order to perform better (Zhang, Hempel, Han, & Tjosvold, 2007). Task interdependence is about exchanging team member’s knowledge and expertise in a non-opportunistic way in order to fulfil assignments. There are several researchers who argue for a positive effect of interdependence on ambidextrous behaviour. For example, Mom et al. (2015) stated that this can be accomplished by building and maintaining a personal network throughout the organization. According to Rogan and Mors (2014) are networks, informal and formal ties, important for managers to behave ambidextrous. There are several reasons to expect the opposite, so that reciprocal task interdependence has a negative effect on ambidexterity. The first reason is that ambidextrous employees need high individual autonomy in order to generate positive returns from the simultaneous pursuit of exploration and exploitation activities (Jansen, Simsek, & Cao, 2012). Langfred (2005) found a negative link between individual autonomy and performance when task interdependence was high, as well as a positive link between individual autonomy and performance when interdependence was low. Since ambidextrous employees need high autonomy in order to pursue ambidextrous goals, reciprocal task interdependence should not be high. The second reason is that this research

14 expects that role ambiguity has a positive impact on the relationship between openness to experience and individual ambidexterity; it would be odd to expect reciprocal task interdependence to have a positive impact as well. If employees are highly interdependent on other colleagues, they preferably would not have ambiguous roles since it would make interdependencies difficult to interpret.

Hypothesis 5: The positive relationship between openness to experience and individual

ambidexterity is negatively moderated by reciprocal task interdependence, so that this relationship is stronger for lower values of reciprocal task interdependence.

Figure 1 displays the different relationships included in the theoretical framework.

Figure 1: A moderated model of the effect of openness to experience on individual ambidexterity

III. Method

Sample and Data Collection

The data collection from this study is obtained via a questionnaire that was filled out during the “Business Strategy Game” (BSG). During the simulation game, teams of 3 to 5 Dutch students were competing against each other. On five different points in time, during the

15 9 weeks that the BSG lasted, the participants were asked to complete a survey that consisted of questions on an individual and team level.

The data from the sample consisted of 584 students who followed a second year bachelor course on strategic management. 117 teams were formed of which 68.1% were male and 31.9 % were female. The age of the students varied between the 17 and 26 year old (Mage=18.53, SDage=1.16). In the first week students were asked to complete a survey of which one part was about the BSG (wave 1) and one part of the survey contained questions about their personality characteristics. Thereafter, the students were asked to fill out a questionnaire every consecutive week (wave 2 to 5). In order to increase the response rate, the participants who completed the survey received a tailor made feedback report on their team performance and implications for improvement. As a consequence the response rates were between the 96 % and 94 %. In total 129 respondents were removed from the dataset because they lagged participation in the survey. The final data set consists of 455 individuals who remained in the study.

The business Strategy Game

The teams were formed by the students themselves. Students who were not able to form a team were randomly assigned to a team by the professors. The teams were tracked and asked to complete a survey at 5 different points in time. The BSG was part of a strategic management course and accounted for 20 % of the total of 6 ECTS. To overcome the possibility of social loafing, 3 % of the student’s grade consisted of peer evaluations. The BSG lasted 9 weeks in which 9 rounds were played in a worldwide athletic footwear industry. During the 9 weeks of the BSG a realistic business environment was created in which the teams encountered business issues that management teams could face as well. The students

16 were told that in the beginning of the game their firms had good revenues and profits and that they were healthy and performing well.

The idea behind the BSG is that it creates a representation of real world situations which are often complex and ambiguous. The teams have to really work as a team in order to perform. Another aspect of the game which makes it a more believable representation of the ‘real’ world is the fact that the participants themselves set the rules of the game. Since the business environment in which the teams operate is in a state of constant flux the teams have to design and unroll a sound internal strategy. The teams have four options with regard to the geographical markets they want to operate in (Asia-Pacific, Europe-Africa, Latin America and North America), and three options with regard to how they want to sell their products (e-commerce, private-label, and wholesale). After each round the competing teams received a performance report on their earnings per share, return on equity, stock price, credit rating and image rating.

Something that needs to be kept in mind concerns the use of students and the generalizability of the study. Despite the fact that behaviour of student teams and professionals are comparable (Clark & Montgomery, 1996; Lant & Montgomery, 1992), the extent to which the results are generalizable needs to be determined with care. The teams of students as within this study can be best compared with project teams. Project teams are only temporarily and set up to complete certain tasks or problems (Morris & Pinto, 2010, p. 70). When a project team is formed the team members have to map and interpret their internal and external environment, plan and carry out activities and redirect when necessary (Mathieu & Rapp, 2009). But as important is to get acquainted with other team members and to resolve potential tensions (Mathieu & Rapp, 2009).

17

Measures

To be entirely complete, an overview of the different items that make up the measures can be found in appendix 1.

Individual ambidexterity was computed using a two-step approach commonly used

by other researchers as well (Mom et al., 2015, 2009). Firstly, exploration and exploitation behaviour of participants were measured using two different scales. Both the exploration as the exploitation scale were adopted from Mom et al. (2009). The five items of exploration contained questions such as: “to what extent did you engage in game-related activities requiring you to learn new skills or knowledge” (1 = rarely, 7 = all of the time). An exploratory factor analysis indicated that one item needed to be dropped. This resulted in a four item scale for exploration with Cronbach’s Alpha scores ranging between α= .61 and α= .73. The five items of exploitation contained questions such as: “to what extent did you engage in game-related activities which you can properly conduct by using your present knowledge” (1 = rarely, 7 = all of the time). An exploratory factor analysis indicated that one item needed to be dropped. This resulted in a four item scale for exploitation with Cronbach’s Alpha scores ranging between α= .60 and α= .71. Subsequently, a joint exploratory factor analysis with all 10 items confirmed that 2 items needed to be removed due to low factor loadings. After running a final exploratory factor analysis all items clearly loaded on their belonging factor with factor loadings above 0.64 with cross-loadings below .19. With eigenvalues greater than 1.9 these results are another indication of discriminant validity. Second, the multiplicative interaction between exploration and exploitation was computed to come up with a scale for individual ambidexterity. As argued by various researchers multiplicative interaction is the most suitable measure for ambidexterity (Cao, Gedajlovic, & Zhang, 2009; Gibson & Birkinshaw, 2004; Mom et al., 2009).

18

Openness to experience is the dependent variable of this study and part of the Big

Five personality traits (Goldberg, 1990,1993). Openness to experiences was measured once at the beginning of the Business Strategy Game since personality traits are stable over time (Goldberg, 1990). Openness to experience was measured using a 10-item questionnaire extracted from the International Personality Item Pool (IPIP) (Goldberg, 1999; Goldberg et al., 2006). Examples of statements to measure openness to experience are “I have a vivid imagination”, “I am full of ideas”, and “I do not have a good imagination” (reversed item) (1 = strongly disagree, 7 = strongly agree). The 10-item questionnaire inquired in the beginning of the BSG in order to capture openness to experience actually captured the Factor V of the Big Five personality traits. The Factor V measures openness to experience, intellect and imagination (Goldberg, 1993). Some questions of the 10-item questionnaire filled out by the respondents in the beginning of the BSG are directly related to openness to experience. An exploratory factor analysis with 10 items confirmed the above stated: 3 scales could be constructed. After removing items with cross loadings, 3 items that properly represent openness to experience loaded on one factor with factor loadings of .46, .68 and .81. Subsequently, the other items were dropped. The Cronbach’s Alpha of the openness to experience measure was α=.61.

Role ambiguity is one of the moderating variables of this study. This study based the

measure of role ambiguity on Rizzo et al. (1970). Role ambiguity was captured by means of five questions such as “I have clear, planned goals and objectives for my job”. An exploratory factor analysis indicated that two items needed to be dropped. After removing the two items, all items clearly loaded on one factor with factor loadings of above.74. This resulted in a three item scale with Cronbach’s Alpha scores ranging between α= .69 and α= .77.

Need for cognition is a construct validated by Cacioppo et al. (1984). The construct

19 have the responsibility of handling a situation that requires a lot of thinking”, “Thinking is not my idea of fun” (reversed coded), and “I really enjoy a task that involves coming up with new solutions to problems” (1= strongly disagree, 7 = strongly agree). An exploratory factor analysis showed dominant factor loadings on one factor: .75, .77 and .81. The Cronbach Alpha score for the need for cognition measure was α= .68.

Creative problem solving was measured by following the study by Gilson and Shalley

(2004). In order to measure creative problem solving in the present study, respondents were asked to answer four questions on creative problem solving such as “I am open to implementation of new ideas and ways of doing things” (1= strongly disagree, 7 = strongly agree). An exploratory factor analysis on the different items of the creative problem solving construct showed a unidimensional factor with factor loadings of above .71. The construct showed a proper reliability rate of α= .72

Reciprocal task interdependence was the last moderating variable of the study.

Reciprocal task interdependence represents the level of knowledge and expertise exchange with colleagues in a non-opportunistic way in order to fulfil assignments. Following Zhang et al. (2007) respondents were asked to answer five questions about reciprocal task interdependence. An example of a statement is “My own performance is dependent on receiving accurate information from others” (1= strongly disagree, 7 = strongly agree). An exploratory factor analysis indicated that one item needed to be dropped, leading to a construct with factor loadings of above .68. The result was a four item scale with a Cronbach’s Alpha score of α= .75.

Control variables

In order to ensure that all results during the analyses were caused by the aforementioned variables, several control variables were included to control for. Firstly, this

20 study controlled for age and gender to make sure that any potential differences between these variables are controlled for. Secondly, this study controlled for nationality. Since the BSG was part of a strategic management course during the bachelor of a European business school a relatively large part of the students was international. To control for any potential differences between participants of different countries (Schwartz, 1999), a dummy variable was created in which Dutch respondents were coded as 1 and non-Dutch respondents were coded as 0.

Measurement model

To check for construct validity of the measures, an exploratory and confirmatory factor analysis containing all items was performed. The Barlett’s test of Sphericity (Bartlett, 1954), to assess the intercorrelations among items, was significant on a 0.001 level (p<0.05) ) and the Kaiser Meyer-Olkin (KMO) measure of sampling adequacy (Kaiser, 1970, 1974) ranged from 0.738 to 0.788 with a suggested minimum of 0.6 (Tabachnick & Fidell, 2007). Overall, the data were suitable for factor analysis. A principal components analysis with Varimax rotation showed that the factors were satisfactorily created. The factor analysis indicated a 7 factor solution with eigenvalues for every factor greater than 1.3. The factor loadings were between the 0.53 and 0.85, but mostly around 0.7 and there were no cross-loadings greater than 0.31. Subsequently, a confirmatory factor analysis (CFA) was conducted using AMOS to validate the findings of the exploratory factor analysis. The proposed model fitted the data well (CFI =.91, TLI=.89, RMSEA=0.04) (Hair, Black, Babin, & Anderson, 2010). The average variance extracted, to measure the variance seized by every construct, ranged between 0.30 and .49 with a threshold of above 0.5 to confirm convergent validity (Hair et al., 2010). This is a limitation that should be taken into account. The composite reliability for the different constructs ranged between .60 and .76 with a suggested minimum

21 of 0.7 (Hair et al., 2010). Lastly, an invariance test showed that the model was invariant. The invariance test suggested that there were no signs of common method bias, and therefore it was not necessary to create a common latent factor.

IV. Results

Table 1 presents the means, standard deviations and intercorrelations of all study variables. In order to assess multicollinearity, the variance inflation factors (VIF) were calculated for each of the regression equations. Given the fact that the highest VIF score was 2.45, which is below the rule of thumb cut-off value of 10 (Neter, Wasserman, & Kutner, 1990), no multicollinearity issues were detected.

Table 1: Means, Standard Deviations and Intercorrelations

Mean SD 1 2 3 4 5 6 7 8 9 1. Individual Ambidexterity 21.67 6.18 - 2. Openness to Experience 3. Role ambiguity 4.93 5.29 0.79 0.85 .24** .44** - .22** - 4. Need for cognition 5. Creative problem solving 6. Reciprocal task interdependence 4.60 5.06 4.08 0.68 0.61 0.70 .13** .28** .28** .44** .28** .13** .18** .42** .26** - .28** .00 - .52** - 7. Age 8. Gender 9. Nationality 18.53 1.33 0.96 1.17 0.47 0.21 -.05 .00 .05 .02 -.02 .01 .03 -.06 -.01 .07 -.12* .09 .06 -.10* .04 -.17** -.05 .14** - -.12* -.13* - .07 -

Note: N=379, Significance legend: *p < .05 and **p < .01

A moderated hierarchical regression analysis was performed to examine the relationship between openness to experience and individual ambidexterity, and the moderating variables: role ambiguity, need for cognition, creative problem solving and reciprocal task interdependence. The independent and moderating variables were introduced in a stepwise process to be able to make comparisons between the different models. Table 2 displays the results of the hierarchical regression analyses including standard errors (SE) and coefficients (β). Model 1 only contains the control variables age, gender and nationality.

22 Model 2 reveals the main effect of openness to experience on individual ambidexterity. Model 3 introduces role ambiguity, need for cognition, creative problem solving and reciprocal task interdependence as independent variables to see whether these variables have an effect on individual ambidexterity. Model 4, 5, 6 and 7 involve the interaction effects between openness to experience and the different moderating variables and individual ambidexterity.

23

Note: N = 378. Significance legend: * p < .05, ** p < .01 and *** p < .001

Individual ambidexterity

Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7 SE β SE β SE β SE β SE β SE β SE β Controls Age Gender Nationality Main effect .28 .68 1.55 -.04 -.04 .05 .27 .66 1.51 -.05 -.03 .05 .25 .61 1.37 -.03 .01 .02 .25 .60 1.37 -.03 .02 .03 .25 .61 1.37 -.03 .02 .03 .25 .60 1.37 -.02 .02 .02 .25 .61 1.37 -.02 .02 .02 Openness to experience Moderation effect Role ambiguity Need for cognition Creative problem solving Reciprocal task interdependence

Interaction effects - - - - - - - - - - .39 - - - - .23*** - - - - .39 .38 .47 .61 .48 .11* .40*** .04 -.04 .18*** .39 .37 .47 .61 .48 .11* .40*** .04 -.04 .18*** .39 .37 .47 .61 .48 .11* .40*** .04 -.04 .18*** .39 .37 .47 .61 .48 .11* .39*** .04 -.04 .19*** .40 .37 .47 .61 .48 .11* .39*** .04 -.04 .19*** Openness to experience x Role ambiguity Openness to experience x Need for cognition Openness to experience x Creative problem solving

- - - - - - - - - - - - - - - - - - .38 - - .09* - - .40 .48 - .09 .01 - .42 .51 .57 .06 -.01 .08 .42 .52 .68 .06 -.02 .10 Openness to experience x Reciprocal task interdependence

- - - .55 -.03 Adjusted R² ∆ Adjusted R² .00 .05 .05*** .24 .19*** .25 .01* .24 .00 .24 .00 .24 .00

24 Hypothesis 1 proposed that openness to experience as a personality characteristic has a positive effect on the ability to simultaneously show exploring and exploiting behaviour (individual ambidexterity). Model 2 in table 2 presents the relationship between openness to experience and individual ambidexterity which is positive and significant (β = 0.23, p < .001). The adjusted R² shows that 5 % (p < .001) of the variance is explained by openness to experience. Therefore, hypothesis 1 is supported.

Although not hypothesized, the moderator variables role ambiguity (β = 0.40, p < .001) and reciprocal task interdependence (β = 0.18, p < .001) added on their own a significantly unique percentage ∆R² =.19 (p < .001) of the variance in individual ambidexterity. This shows that 19 % of individual ambidexterity is explained by the variables role ambiguity and reciprocal task interdependence. Subsequently, in order to analyze the moderation effects, the moderator variables and the independent variable were mean centered before creating the interaction terms.

Hypothesis 2 predicted a positive moderating effect of role ambiguity on the relationship between openness to experience and individual ambidexterity. Model 4 in table 2 presents the interaction coefficient which is positive and significant (β = 0.09, p < .05). For a more detailed inspection of this significant interaction effect the simple moderation model from the PROCESS macro for SPSS was used (Hayes, 2013). The analysis showed the conditional effects of openness to experience on individual ambidexterity at different values of role ambiguity (low, mean and high) (+/- 1 SD below mean). The results showed that at low values of role ambiguity there was a non-significant relationship between openness to experience and individual ambidexterity (t = 0.44, p > .05). At a mean value for role ambiguity there was a positive significant relationship (t = 2.14, p < .05), and at high values for role ambiguity there was also a positive significant relationship between openness to experience and individual ambidexterity (t = 2.91, p < .05). The output of the

Johnson-25 Neyman technique that shows the significance regions can be found in appendix 2. The plot of this interaction is shown in figure 2. It shows a positive relationship between openness to experience and individual ambidexterity when role ambiguity is high. Hypothesis 2 is therefore supported.

Figure 2: The interaction between openness to experience and role ambiguity on an

individual’s ambidexterity.

Hypothesis 3 predicted a positive moderating effect of need for cognition on the relationship between openness to experience and individual ambidexterity. Model 5 in table 2 introduces the interaction effect which is positive but not significant (ß = .01, n.s.). Therefore, hypothesis 3 is not supported by the analysis.

Hypothesis 4 predicted a positive moderating effect of creative problem solving on the relationship between openness to experience and individual ambidexterity. Model 6 in table 2 introduces the interaction effect which is positive but not significant (ß = .08, n.s.). Therefore, hypothesis 4 is not supported.

Hypothesis 5 predicted a negative moderating effect of reciprocal task interdependence on the relationship between openness to experience and individual

26 ambidexterity. Model 7 in table 2 introduces the interaction effect which is negative but not significant (ß = -.03, n.s.). Although there is a main effect of openness to experience on individual ambidexterity and a main effect of reciprocal task interdependence on individual ambidexterity, when creating an interaction term there is no significant effect. Therefore, no support was found for hypothesis 5.

V. Discussion

The objective of this study was to continue and deepen research on individual ambidexterity since research within this area is still underdeveloped (Lavie et al., 2010; Raisch et al., 2009). Ambidexterity is the ability to simultaneously pursue two seemingly contradicting concepts namely, exploration and exploitation. Thus, ambidexterity is not an additional personality characteristic or trait (Bledow, Frese, Anderson, Erez, & Farr, 2009). Mom et al. (2007) showed that some individuals were able to show high levels of exploration and exploitation activities, making them ambidextrous. These individuals are important for organizations that aim to pursue ambidextrous goals. Research has already shown that ambidexterity leads to higher performance of a firm, business unit or an individual (O’Reilly & Tushman, 2013). This means that organizations search for individuals that are able to act ambidextrous. In order to be able to select individuals on their ability to behave ambidextrous it is of key importance to know why some individuals are better able to behave ambidextrous than others. Various researchers tried to answer this question by searching for conditions and characteristics under which individuals behave ambidextrous (Good & Michel, 2013; Keller & Weibler, 2014; Mom et al., 2015, 2009). The current study tried to expand the knowledge on individual ambidexterity by taking a trait approach. Thereby also answering to the call to explain why some individuals are better able to act ambidextrous than others. This research especially focussed on the role of openness to experience as a predictor for ambidextrous

27 behaviour. Drawing on previous research it was hypothesized and confirmed that openness to experience is positively related to individual ambidexterity (Bonesso et al., 2014; Gupta et al., 2006; Keller & Weibler, 2014; Zacher et al., 2014). Individuals who are open to new experiences are better able to act ambidextrous. Although this is the first research to examine the relationship of openness to experience and individual ambidexterity, this result is in line with comparable previous findings. For example, Keller and Weibler (2014) found a negative moderating effect of openness to experience on the relationship of individual ambidexterity and cognitive strain. This already implied that individuals who score high on openness to experience are better able to cope with the tensions that arise from simultaneously pursuing exploration and exploitation related activities. Good and Michel (2013) state that individual ambidexterity is always based on intelligence. Since intelligence is positively correlated to openness to experience (Aitken Harris, 2004), it is a logical expectation that openness to experience is positively associated with individual ambidexterity. A study by Zacher et al. (2014) found that openness to experience is a strong predictor for exploration as well as exploitation. This is another comparable outcome that points to a positive relation between openness to experience and individual ambidexterity.

Subsequently, a research stream on situational specificity argued that individuals do not live in a vacuum, so that their choices depend on the context they operate in (Gibson & Birkinshaw, 2004; Morgeson & Hofmann, 1999; Simsek, 2009). Following this line of reasoning role ambiguity, need for cognition, creative problem solving and reciprocal task interdependence were expected to moderate the relationship between openness to experience and individual ambidexterity. Surprisingly, this study showed that only role ambiguity had a significant influence on the relationship between openness to experience and individual ambidexterity. So that the relationship between openness to experience and individual ambidexterity is stronger for higher values of role ambiguity.

28 This moderating relationship makes sense if one looks at the factors that make up the openness to experience trait. Individuals who are among others creative, imaginative, intellectual and liberal would seem comfortable under uncertain conditions. They would perform better under uncertain conditions since they are able to combine a long- and short-term orientation, using internal and external knowledge (Mom et al., 2015). They would experience strain in roles with high clarity, since there would be no room to explore new opportunities and experiences. Mom et al. (2015) add that ambidextrous individuals should be able to utilize a variety of possible roles.

Need for cognition, reciprocal task interdependence and creative problem solving did not have a significant influence on the relationship between openness to experience and individual ambidexterity. These results, contradicting the hypotheses, were somewhat surprising for multiple reasons. Firstly, it was expected that need for cognition had a positive moderating effect on the main relationship, because people who score high on need for cognition prefer working in complex uncertain situations, are good learners, and are convincing in transferring their ideas (Wu et al., 2014). These are typically characteristics that fit an ambidextrous individual. A possible explanation for the non-significant finding might be the fact that the relationship between need for cognition and exploration behaviour is lower for individuals who have high autonomy jobs (Wu et al., 2014). One can argue that the respondents of this study had high autonomy jobs during the BSG, because their teams were independent and they were allowed to set the rules of the game by themselves. Since ambidexterity consists of half exploration behaviour and autonomy is a condition for ambidexterity (Chang & Hughes, 2012; Raisch & Birkinshaw, 2008; Raisch et al., 2009), this might be a reason for the non-significant finding of the moderating role of need for cognition. Secondly, it was expected, but not confirmed, that creative problem solving had a positive moderating effect on the main relationship. This means that the relationship between

29 openness to experience and individual ambidexterity is not stronger for higher values of creative problem solving. One reason might be that although openness to experience is related to creativity (George & Zhou, 2001; McCrae, 1987; Taggar, 2002), individual ambidexterity has no direct relation to creativity.

Lastly, it was expected that reciprocal task interdependence had a negative moderating effect on the main relationship of this study. This means that the relationship between openness to experience and individual ambidexterity is stronger for lower values of reciprocal task interdependence. Although this study expected a negative moderating effect, there were also arguments for a positive moderating effect. For example Rogan and Mors (2014) stated that informal and formal ties are important in order to behave ambidextrous. One of the arguments in favour of a negative moderating effect was made by Langfred (2005), who found a negative link between individual autonomy and performance when task interdependence was high. Langfred (2005) also found a positive link between individual autonomy and performance when interdependence is low. As stated earlier, autonomy is a precondition for ambidexterity which, according to Langfred (2005), would automatically imply low values for reciprocal task interdependence. As a result a positive moderating effect as well as a negative moderating effect was arguable, but it is not exactly clear why reciprocal task interdependence as a moderator is non-significant in both directions. This study did however find a positive direct effect of reciprocal task interdependence on individual ambidexterity. This implies that individuals who exchange knowledge and expertise with other team members or colleagues are better able to behave ambidextrous. Which is in line with the findings of Mom et al. (2007) who showed that top-down knowledge inflows had a positive effect on manager’s exploitation behaviour and bottom-up and horizontal knowledge inflows had a positive effect on manager’s exploration behaviour.

30 According to the results of this study and the interactionist model of Tett and Burnett (2003) the personality characteristic openness to experience would only get expressed if someone would have a high ambiguity in their role. Role ambiguity, which was on a social level of trait relevant cues, would according to Tett and Burnett (2003) lead to the expression of the trait openness to experience. Need for cognition, creative problem solving and reciprocal task interdependence, which were on a task and social level of cues do not lead to the expression of openness to experience. In other words, these variables do not have significantly moderating effect on the relationship between openness to experience and individual ambidexterity.

Theoretical and practical implications

This study offers several contributions to the current body of knowledge on ambidexterity. The first contribution is made by further incorporating personality characteristics into the ambidexterity literature. Since it is still unclear why some individuals are better able to act ambidextrous than others, prior research encouraged the incorporation of personality characteristics (Raisch et al., 2009). By taking a trait approach, this study aimed at explaining ambidextrous behaviour from a yet underexplored perspective. Although openness to experience is just one of the many personality characteristics that might be able to explain why some individuals are better able to act ambidextrous, the significant result does confirm that personality characteristics play an important role. If in the future research would be able to compose a complete overview of characteristics that account for ambidextrous behaviour, it could be used for allocating an employee to tasks which require individual ambidexterity.

The second contribution is made by answering to the call to explore the role of context when investigating individual ambidexterity. Several prior researchers pleaded for the incorporation of context into the individual ambidexterity literature (Good & Michel, 2013; Mom et al., 2015; Raisch et al., 2009). This study incorporated several contextual moderators

31 to test whether they had an impact on the relationship between openness to experience and individual ambidexterity. Role ambiguity was the only moderator that significantly influenced the main relationship. This means that individuals who are open to new experiences prefer working in roles that are ambiguous. Ambiguous in a sense that for example time and resources are not allocated in advance.

An important practical implication for managers is to know how ambidextrous managers can be utilized in their very best way. One way of doing this is by making use of the framework by Floyd & Lane (2000). Their framework about strategic renewal consists of three sub processes namely: competence definition, competence deployment and competence modification. During the competence modification process the need for strategic change is recognized by managers. On the one hand, the current business still needs to be exploited in order to maintain viability. On the other hand, management is looking for new opportunities to explore. Due to the recognition for change, control systems are relaxed and mutual adjustments are encouraged (Floyd & Lane, 2000). Since the competence modification process requires a high tolerance for ambiguity and flexibility in decision making (Floyd & Lane, 2000). This is where especially ambidextrous managers have added value.

Limitations and future research

This study has several limitations that should be taken into account, but do offer possibilities for future research. The first limitation concerns the context in which the data of this study was collected. The setting was a Business Strategy Game with Dutch undergraduate students as part of a management course, which makes the sample non-random. Participation was encouraged by means of a customized feedback report informing them how to improve performance. Subsequently, 20 % of their grade for the management course was composed of the team performance. This might have motivated the students to participate, leading to a high

32 response rate. On the other hand the length of the business game and the amount of questionnaires that needed to be completed may have led to less complete and serious responses. If there is a significant amount of respondents who were less serious when participating in the surveys because they only wanted a customized report on team performance it would jeopardize the value of the results of the study. For future research it would be valuable to conduct a similar study in a real life business setting.

The second limitation concerns how openness to experience was measured. Openness to experience was the independent variable and part of the Big Five personality traits (Goldberg, 1990, 1993). Openness to experience was measured in the first week as part of the questionnaire about personality characteristics. Based on Goldberg’s Big Five Personality Score, 10 7-item questions were used to measure openness to experience (Goldberg, 1999; Goldberg et al., 2006). A complication is that these 10 questions not only measured openness to experience, but also intellect and imagination. This might be attributed to the fact that the questionnaire tried to capture Factor V of the Big Five domain and not only openness to experience. Goldberg (1993) concluded that Factor V is not even properly explained by openness to experience and intellect and maybe even better by imagination. This means that the questionnaire used for this study measured only a part of the three personality characteristics. As a consequence only three of ten items were used in the analysis to explain openness to experience. This still gives a proper explanation about the relationship between openness to experience and individual ambidexterity, but the relationship might have been even stronger with a more complete measure of openness to experience. For further research it would be a valuable contribution to use a full measure of openness to experience (Costa Jr. & McCrae, 1995). In this way the relationship between openness to experience and individual ambidexterity can be more fully explained.

33 The third limitation concerns the validity issues that arose during the confirmatory factor analysis. Almost all of the values for average variance extracted (AVE) were below the threshold of 0.5 (Hair et al., 2010). This convergent validity issue means that there are too little correlations of the items within the different factors, so that the items that make up the factors would not properly explain them. Although this was not expected, since the questionnaire consisted entirely of validated questions, it is still a limitation that should be taken into account.

Another promising avenue for future research is to incorporate more personality characteristics in the individual ambidexterity research. In order to find out why certain individuals are better able to behave ambidextrous, with the presumption that openness to experience is significantly important, other personality characteristics might be evenly or more meaningful.

VI. Conclusions

This study examined the relationship between openness to experience and individual ambidexterity. Results indicate that openness to experience is positively related to individual ambidexterity. Individuals who are more open to experiences are better able to behave ambidextrous. Subsequently, this study assessed if there are contextual conditions under which the main relationship is even stronger. The analysis indicated that role ambiguity has a positive effect on the relationship between openness to experience and individual ambidexterity. Results also showed that need for cognition, creative problem solving and reciprocal task interdependence did not have a significant effect on the relationship between openness to experience and individual ambidexterity. Surprisingly, the analysis revealed that reciprocal task interdependence did have a direct significantly positive effect on individual’s ambidexterity. This means that individuals operating in a work context in which their

34 dependent on other colleagues are better able to act ambidextrous. This study took an important step in the investigation of personality characteristics as a predictor for ambidextrous behaviour. The most important question was and is why some individuals are better able to act ambidextrous than others. Future research may find other personality characteristics that are equally or even more meaningful in explaining this question.

35

VII. References

Aitken Harris, J. (2004). Measured intelligence, achievement, openness to experience, and creativity. Personality and Individual Differences, 36(4), 913–929. http://doi.org/10.1016/S0191-8869(03)00161-2

Barrick, M. R., & Mount, M. K. (1991). The Big Five Personality Dimensions and Job Performance: A Meta-Analysis. Personnel Psychology, 44(1), 1–26.

Barrick, M. R., Mount, M. K., & Judge, T. A. (2001). Personality and Performance at the Beginning of the New Millennium: What Do We Know and Where Do We Go Next?

International Journal of Selection and Assessment, 9(1-2), 9–30. http://doi.org/10.1111/1468-2389.00160

Bartlett, M. S. (1954). A Note on the Multiplying Factors for Various χ2 Approximations.

Journal of the Royal Statistical Society. Series B (Methodological), 16(2), 296–298.

Bledow, R., Frese, M., Anderson, N., Erez, M., & Farr, J. (2009). A Dialectic Perspective on Innovation: Conflicting Demands, Multiple Pathways, and Ambidexterity. Industrial

and Organizational Psychology, 2(3), 305–337.

http://doi.org/10.1111/j.1754-9434.2009.01154.x

Bonesso, S., Gerli, F., & Scapolan, A. (2014). The individual side of ambidexterity: Do individuals’ perceptions match actual behaviors in reconciling the exploration and exploitation trade-off? European Management Journal, 32(3), 392–405. http://doi.org/10.1016/j.emj.2013.07.003

Borman, W. C., & Brush, D. H. (1993). More Progress Toward a Taxonomy of Managerial Performance Requirements. Human Performance, 6(1), 1–21. http://doi.org/10.1207/s15327043hup0601_1

Briñol, P., & Petty, R. E. (2005). Individual differences in attitude change. Mahwah, NJ: Lawrence Erlbaum.