1
Autonomy: a new pathway to
contextual ambidexterity.
A survey research
BSC Business Administration
Student name: Ifigenia Pansters Student number: 11872586 Supervisor: Eugène Hoogstad Date: 22nd of June 2020
2 Statement of originality
This document is written by Ifigenia Pansters who declares full responsibility for the content of this document. I declare that the work presented in this document is original and that no sources other than those mentioned in the text and references have been used in creating it. The faculty of Economics and Business is responsible solely for supervision of completion of work.
3 Table of Contents Abstract ... 4 Introduction ... 5 Literature review ... 7 Autonomy ... 7
Autonomy: unidimensional or multidimensional ... 7
Strategic versus structural autonomy ... 9
Ambidexterity ... 10
Temporal ambidexterity ... 11
Structural ambidexterity... 11
Contextual ambidexterity ... 12
Contextual Ambidexterity and autonomy ... 13
Theoretical framework ... 15
Structural autonomy and exploitation ... 15
The moderating role of need for autonomy... 16
Strategic autonomy and exploration ... 16
Moderating role of creative self-efficacy ... 17
Contextual ambidexterity ... 18 Methods ... 19 Sample ... 19 Measures ... 19 Analytical Plan... 21 Results ... 22
Results linear regressions ... 23
Results moderations ... 24
Discussion... 27
Limitations and further research ... 29
Managerial implications ... 30
Conclusion ... 30
Reference list ... 31
4 Abstract
We are currently living in a highly uncertain world where being innovative is crucial for businesses to survive. Unfortunately, many companies today are still struggling with exploiting existing competences and at the same time with the ability to explore competences beyond the direct scope of their business. The ability to master both of these activities is known as ambidexterity. This paper suggests that in the case of start-up company’s autonomy has a positive effect on ambidexterity. More specifically, I hypothesize that structural and strategic autonomy will lead to the increased formation of contextual ambidexterity. Moreover, I expect that this relationship is stronger when employees have a high need for autonomy and creative self-efficacy. I tested these hypotheses with a sample of 60
respondents. The hypothesis that strategic and structural autonomy would have a positive influence on exploration and exploitation and consequently contextual ambidexterity was supported. This indeed suggests that providing employees with more strategic and structural autonomy causes them to perform more explorative and exploitative activities thereby increasing the formation of contextual ambidexterity. However, the need for autonomy and creative self-efficacy amongst employees does not influence this relationship.
5 Introduction
Whether you are a CEO, a business student, or a hairdresser, you have, without a doubt, experienced the greatly uncertain and volatile world we live in today. In 2020 we experience a global pandemic that has shaken not only consumers' daily lives but the entire business world. Companies, small or large, needed to adapt quickly and come up with innovative ways to remain afloat. The Covid-19 pandemic constitutes an existential threat for many
companies. However, it is not only pandemics that can condemn firms into obsolescence. Technological change accompanied by rapid innovation can also become significant threats. They create a highly volatile world that increasingly requires organisations to gain the ability to exploit existing competences and at the same time explore new one’s competences (Wang & Rafiq, 2014). The balancing of both these activities to achieve successful innovations and thereby ensure survival has been referred to as organisational ambidexterity.
The ambidexterity literature identifies several models to create ambidexterity within firms. Initially, the advised model to implement was that of structural ambidexterity, whereby separate units are created for exploration and exploitation (Wang & Rafiq, 2014). This was seen as the optimal option as these exploration and exploitation activities were perceived to be paradoxical and could not exist together under the same culture and processes (Wang & Rafiq, 2014). Exploration for example, was expected to occur in small and
decentralised units with loose processes facilitating experimentation, flexibility and divergent thinking (Raisch & Birkinshaw, 2008) (Jansen et al., 2009). In contrast, exploitation occurs in larger, centralized units with tight processes and a focus on efficiency and refinement of production (Raisch & Birkinshaw, 2008) (Jansen et al., 2009). In other words, structural ambidexterity enables ‘pragmatic boundaries’ that safeguard the differences between the exploration and exploitation unit (Jansen et al., 2009).
However, other studies have emphasized contextual ambidexterity, a model whereby individuals pursue exploration and exploitation simultaneously in the same business unit (O’Reilly & Tushman, 2013). According to Wang and Rafiq (2014) this form of
ambidexterity helps resolve some coordination costs that arise from the structural approach as it characterises the development of the organisation towards the integration of both activities. It can be seen as a bottom-up approach that stresses the involvement and participation of individual employees (Wang & Rafiq,2014).
For small to medium-sized firms to have differentiated units for exploration and exploitation is more difficult since they exist in a resource-constrained context (Hansen, Wicki & Schaltegger, 2018). As exploration and exploitation compete for scarce resources,
6
organisational routines and attention, the situation of limited resources is probable to amplify the paradoxes faced. As a consequence, structural ambidexterity is often not possible because of the large resources and costs it requires. Therefore Volery et al. (2013) identify contextual ambidexterity as especially relevant to SME's. Because I want to focus on the topic of ambidexterity in start-up companies, the notion of contextual ambidexterity will be the central point of attention of this research project.
While the benefits of achieving contextual ambidexterity for performance and innovation are widely known, many questions surrounding the management practices and culture that create this are still left unanswered. This study would like to contribute to this gap in the literature by providing a management practice that can be implemented to boost exploration and exploitation. I will pay specific attention to the implementation of higher levels of autonomy amongst employees.
Autonomy is a very broad concept with a wide variety of interpretations. The key distinction made in this paper is between structural and strategic autonomy. The background and debates within the field are examined in the literature review. Structural autonomy refers to the extent in which employees enjoy freedom regarding factors within the direct work environment (Lumpkin, 2009). Strategic autonomy relates to the freedom regarding goal setting and thus its planning and conducting (Lumpkin, 2009). This paper posits of
hypothesizes that both forms of autonomy have a significant impact on either exploration or exploitation. This is further studied in the theoretical framework.
To conclude, the purpose of the study is to test if autonomy can facilitate contextual ambidexterity. My focus is on start-ups. I therefore hope to contribute to a literature that has overwhelmingly focused on ambidexterity within large firms. Research into this effect within start-ups is essential as these can also greatly benefit from ambidexterity (Sinha, 2015). Moreover, I have opted to work with online start-ups as they are particularly sensitive to the volatility resulting from technological innovations. The paper consists of, firstly, an extensive literature review and theoretical framework. The following sections deal with methods, measures, my analytical plan and the main results of this study. Finally, there will be a discussion on the findings and a mention of limitations and implications for further research.
7 Literature review
In the literature review, I will provide the reader with an extensive overview of the current debates and findings in the field of study about autonomy and ambidexterity. First, we will look at the concept of autonomy, followed by ambidexterity. I conclude this section with how this feed into the central research question of the paper.
Autonomy
The positive effects of autonomy on employee involvement and skills development have been widely demonstrated (Lopes, Calapez & Lopes, 2015). Moreover, studies have indicated that it leads to higher job performance and job satisfaction (Lopes, Lagoa & Calapez,2012). These benefits have led autonomy to become a popular concept amongst researchers. Over time, it has been studied and conceptualized in various ways (Breaugh, 1999). To give a good overview of the literature this paper will discuss two distinctions that have been made. First of all, there will be a review of the discussion on work autonomy as a unidimensional or multidimensional construct. Secondly, I will look at the distinction between structural autonomy that has been elaborated in the job design literature and the notion of strategic autonomy that has been emphasized in the entrepreneurial literature.
Autonomy: unidimensional or multidimensional
During the 1970s and 1980s, researchers focused primarily on autonomy as a global concept which could be measured on a unidimensional scale (Breaugh, 1999). A widely used example of this conceptualisation is that of Hackman and Oldham (1975, in Sadler-Smith, El-Kot & Leat, 2003), who incorporated autonomy as one of the several core job design
characteristics in their job design theory. In this conceptualisation, work autonomy was defined as 'the degree to which a job provides substantial freedom, independence and discretion to the individual in scheduling and determining the procedures to be used in carrying it out' (Hackman & Oldham, 1975 in Sadler-Smith et al., p.710). After this initial conceptualization, in the 1980s a scholarly discussion emerged about whether job autonomy was in fact best understood as a unidimensional concept.
One of the leading researchers was Kiggundu, who in 1983 (in Breaugh, 1985)
documented the value of distinguishing facets of the notion of autonomy instead of viewing it as a global concept. De Cotiiss and Koys (1980, in Breaugh, 1985, p. 555) corroborated this view by defining autonomy as “the perception of self-determination with respect to work procedures, goals and priorities”. This definition proposes a differentiation of three facets
8
namely; procedures, goals and priorities. This was further emphasized by Nicholson (1984, in Breaugh 1985) who underscored the importance of these three facets within the global
concept of autonomy. Considering all this research, Breaugh (1985) concludes that autonomy can and should be operationalized into three facets. Defined by Breaugh (1985, p.556) these are: (1) method autonomy, the degree to which individuals have a choice regarding
procedures or methods they use; (2) scheduling autonomy, referring to the extent to which employees feel they have the ability to control scheduling, sequencing and timing of work activities; and (3) criteria autonomy, meaning the degree to which employees can choose the criteria on which their work is evaluated.
Despite these conceptual steps, no instrument had been developed that could measure these facets separately. As a consequence, research continued to use the scale originally developed by Hackman & Oldham (1975, in Breaugh, 1999) that measured global autonomy. To be able to further the research in the field of autonomy, Breaugh (1985) then set out to develop an instrument to measure each of facets independently. This new scale was validated by Breaugh (1985) and later by Breaugh (1999). A more recent study By Sadler-Smith, El-Kot and Leat (2003) also set out to test the hypothesis that autonomy could be divided into three facets. Their results confirmed autonomy as a multidimensional concept and that this conceptualization is valid outside the context from which it was originally derived (in Sadler-Smith, El-Kot & Leat, 2003, p.727).
However, not all studies led to the same outcome. A study conducted by Brady et al. as mentioned by Sadler-Smith et al. (2003) had not found significant evidence for the existence of the before-mentioned facets. The mixed results and views concerning the question to measure autonomy as a global or a multidimensional concept have led to a great differentiation within literature. There are two variations within the field of autonomy that are worth mentioning. The first is regarding studies that, contrary to Breaugh (1985), continue to measure autonomy as a global measure, as it was first introduced by Hackman and Oldham (1975). Research done by Burcharth, Præst Knudsen and Søndergaard (2017) sets out to investigate the role of employee autonomy on open innovation performance. To measure employee autonomy, Burcharth et al. (2016) use a global measurement scale. This
exemplifies that even more recent studies still conceptualize autonomy as a unidimensional construct.
The second variation falls within the multidimensional domain of autonomy. An example is the study done by Lopes et al. (2012). While years ago, Breaugh (1985) identified three facets, Lopes et al. (2012) only identify two. The first being procedural autonomy, the
9
control workers have over their methods and schedules, whilst content autonomy concerns the control workers have over their work content and use of their skills. The two facets identified by Lopes et al. (2012) are comparable to that of Breaugh (1985) as it also identifies the freedom of methods and scheduling. However, the difference is that the final
conceptualisation of the facets has been altered.
In short, there has been interesting scholarly debate about if the measurement of autonomy should be on a unidimensional or multidimensional scale. For this particular study, the multidimensional scale by Breaugh (1985) will be used as it is the most comprehensive and inclusive scale.
Strategic versus structural autonomy
Within the field of autonomy, different classifications have also been developed. This can be seen in the job design literature and in the entrepreneurial literature. As identified by Lumpkin (2009) there is a distinction between structural autonomy and strategic autonomy. The former refers to the extent to which freedom is granted regarding factors within the work environment (Lumpkin, 2009). According to Gard et al. (2013) this concerns the ability to define work criteria, scheduling and work methods without the need to seek approval. We see in this conceptualisation a comparison to the earlier mentioned facets from Breaugh (1985) which were; criteria scheduling and methods autonomy (Breaugh, 1985). One could thus argue that definition of Gard et al. (2013) of structural autonomy is the same as the autonomy identified by Breaugh (1985). This view is corroborated by Lumpkin (2009) who also sees Breaugh’s work (1985) as an example of structural autonomy. Thus, the discussion on the multidimensional scale falls under the broader term structural autonomy.
However, autonomy from an entrepreneurial orientation refers not to structural but strategic autonomy (Lumpkin, 2009), which can be defined as the extent to which a group has control over its goals as well as the degree of independence employees have to choose the problems they work on (Lumpkin, 2009) (Globocnik & Salomo, 2014). In contrast to structural autonomy, which falls within the previously established boundaries of the job, strategic autonomy enables employees to explore business opportunities outside the chain of command and the current strategy of the firm (Corsino, Giuri & Torrisi, 2018). In other words, it gives workers the freedom to explore possibly innovative activities without approval even if they obui8find themselves outside the firm’s strategic scope (Burcharth et al., 2017).
10
This form of autonomy stimulates employees to elaborate their ideas outside official communication channels and encourages employees to engage in risk taking that is expected to favour new business developments (Gard et al., 2013) (Globocnik & Salomo, 2014). As a consequence, it nurtures the development of an entrepreneurial orientation (Corsino et al., 2018).
In conclusion, we have identified two key differences between structural and strategic autonomy. The first concerns the kind of freedom granted: structural autonomy is in regard to freedom within confinements of the job, while strategic autonomy concerns the control employees have over their goals and problems they work on (Lumpkin, 2009). Secondly, the effect each form has differs greatly. Strategic autonomy can lead to the discovery of new opportunities that do not fit with the current scope of the organisation, whilst structural autonomy is likely to lead to the development of innovations within the existing scope (Corsino et al., 2018). The latter is fundamental to the purpose of this paper and will be further elaborated in my theoretical framework.
Ambidexterity
Organisational ambidexterity can be defined as a firm’s ability to explore new competencies but also to exploit existing ones (Wang & Rafiq,2004). Exploration can be defined as the discovery of new products, resources and opportunities for new businesses and different ways of doing business (Sinha, 2015) (Chen, 2017). Explorative behaviour is
characterised by discovery, experimentation, risk taking and radical innovation (He & Wong, 2004). Inversely, exploitation focuses on what a firm already knows and concerns the
possible refinement or improvement of existing products, resources and competencies (Sinha, 2015) (Chen, 2017). It is characterised by incremental changes, implementation, efficiency, production and selection (He & Wong, 2004) (Sinha,2015).
Since, exploration and exploitation are seen operating according to fundamentally different logics, it is likely that tensions between them are likely to emerge (He & Wong, 2004). After all, they compete for scarce resources and managers are confronted with trade-off decisions (He & Wong, 2004). Initially, exploration and exploitation were seen as opposite activities and their balance could only be achieved by separating them (Wang & Rafiq, 2014). However, other studies have shown that it is possible to have simultaneous performance in both activities within the same business unit (Wang & Rafiq,2014).
This shows that there are several frameworks on how ambidexterity can possibly be conceived and organised within organisations. Taking into the consideration the constraints
11
of this study, only three designs of ambidexterity will be discussed: structural, contextual and temporal ambidexterity. These designs were chosen as they have been most extensively investigated over the past 15 years (O’Reilly III, Tushman, 2013). It is important to clarify that whereas contextual and structural are identified as simultaneously operating
ambidexterity, the case of temporal ambidexterity is conceived as a sequential process (Sinha, 2015).
Temporal ambidexterity
In temporal ambidexterity exploration and exploitation do exist in the same
organisational unit (Volery, Meuler & Siemens, 2013). However, they take place at different points in time to create temporal cycles with a long period of exploitation and short bursts of exploration (Volery et al., 2013). Argumentation for this process is that it is easier for firms to switch between their formal structures than it is to change their entire culture and informal context (O'Reilly III, Tushman, 2013). Moreover, it enables organisations to use different managerial practices when working on projects at different stages (Chen, 2017). Studies have shown that sequential ambidexterity is a good fit for stable more slow-moving markets (O'Reilly III, Tushman, 2013). However, there are also several negative effects related to the implementation of this model. First of all, unit managers still need to judge how to best divide up time and workgroups so as to meet the requirements for exploitation or exploration alternatively (Gibson & Birkinshaw 2004). Second, it is important to keep in mind that the success of this type of ambidexterity depends on the ability to manage conflicts, to develop mechanisms and preserve effective interpersonal relations that assist the switch between activities (Wang & Rafiq,2014). Furthermore, the switching between exploration and exploitation and the reconfiguration it entails can lead to dislocation in the firm and might destroy core organisational capabilities (Chen,2017).
Structural ambidexterity
As mentioned before, the dual requirements of exploration and exploitation were seen unable to co-exist (Wang & Rafiq,2014). Therefore, to achieve ambidexterity it was
necessary to separate exploration and exploitation (O'Reilly III, Tushman & 2004). The answer, therefore, was structural ambidexterity, creating separate structures for different types of activities (Birkinshaw & Gibson, 2004). Within an organisation, there should be a dual structure (Gibson & Birkinshaw, 2004). Within these separate units, there would be freedom to create new processes, structures and culture (O'Reilly III, Tushman & 2004).
12
However, the separate units require to be held together by an overarching strategy and value system (O'Reilly III, Tushman 2013). The benefit of this perspective is that separation enables each unit to be configured to its specific needs of the task environment (Gibson & Birkinshaw, 2004). This often means that the units pursuing exploitation are centralised, are large and contain fewer flexible processes (Sinha, 2015). Alternatively, exploration units are much smaller and have loose processes and are generally more decentralised.
The main downside of structural ambidexterity is that same separation of activities (Birkinshaw, 2004). It is not difficult to see that often ideas from the exploration unit fail because few linkages exist to the core business (Birkinshaw, 2004). Having structurally differentiated organizational units does not necessarily ensure that the successful pursuit of both activities and their mutual reinforcement occur (Jansen et al., 2009). To become a structural ambidextrous organisation, integration mechanisms need to be in place that coordinate and integrate operational capabilities at different locations. This is a necessary step to capture value and achieve effective ambidexterity (Jansen et al., 2009).
Contextual ambidexterity
The last form of ambidexterity is contextual ambidexterity. Within this concept exploration and exploitation are complementary activities that take place simultaneously in the business unit (Wang & Rafiq, 2014). In other words, it emphasizes the integration of both activities within a unit and allows for a differentiated effort in both (Wang & Rafiq, 2014). Individual employees make their own choices on explorative and exploitative-oriented activities in their day-to-day-work (Birkinshaw & Gibson, 2004). Unlike structural and temporal ambidexterity where managers are in charge, this modality allows employees to make their own judgement on how to best divide their time between the two activities (Birkinshaw & Gibson, 2004). Contextual ambidexterity can be viewed as a meta-level capacity for exploration and exploitation that invades all functions and levels in a unit across the organisation (Gibson & Birkinshaw, 2004). It is believed to be more sustainable than the structural separation model as it enables the adaptation of explorative activities across the entire unit, not only the separated functions that were responsible for business development (Gibson & Birkinshaw, 2004). As a consequence, the entire firm is geared towards the integration of exploration and exploitation through the process of organisational learning (Wang & Rafiq, 2014). It also avoids coordination costs (between separated units), which arise in the structural ambidexterity model, and transaction costs (of switching between stages), which feature in the temporal ambidexterity model (Wang & Rafiq, 2014). One of
13
the downsides of this model is the assumption that a single organisational context can (easily) facilitate both exploration and exploitation (Chen,2017). After all, exploration exploitation tends to prosper in different contexts (Chen, 2017). Therefore, contextual ambidexterity requires not only organisational slack resources to allow these exploratory activities, but it also needs the corresponding effective mechanisms that integrate both (Wang & Rafiq, 2014).
Contextual Ambidexterity and autonomy
Now that we have analysed the two core variables of this study, we need to look at the relationship between them. For contextual ambidexterity to succeed its components need to be intertwined in the operational and strategic activities of a business unit (Wang & Rafiq, 2014). To achieve this, individuals must have the ability to make their judgement on how to divide their time between exploratory and exploitative activities and how to integrate them (Wang & Rafiq, 2014). In other words, a bottom-up approach is required, emphasizing the perspectives and practices of employees (Wang & Rafiq, 2014). The focus should be on individuals within organisational systems and processes that support the activity in exploration and exploitation (O'Reilly III, Tushman, 2013). What we can see is that for contextual ambidexterity to succeed individuals need to make their own judgements and have the freedom to do this. As argued before, structural and strategic autonomy assumes
employees this required freedom. In this context, Gibson and Birkinshaw (2004) have made an important contribution. They have identified four elements in the organisational context that facilitate contextual ambidexterity; stretch, discipline, support and trust. More autonomy for employees is expected to increase support which in turn is key for building contextual ambidexterity (Sinha, 2015).
In conclusion, my review of the literature on ambidexterity and autonomy gives insight into what the variables mean and what the possible relations among them could be. Taken all this into consideration, this paper suggests the following research question: does structural and strategic autonomy lead to higher levels of exploration and exploitation
practices which together form contextual ambidexterity and is this relationship stronger when employees have a high need for autonomy and creative self-efficacy within start-ups? The following section will give a detailed oversight of the hypothesis and the literature that flows from this. Below the proposed relationships are shown as conceptual models.
14 Conceptual Model 1 Conceptual Model 2 Conceptual Model 3 Need for autonomy Structural autonomy Exploitation Creative self-efficacy Strategic autonomy Exploration Exploitation Contextual ambidexterity Exploration
15 Theoretical framework
This section will elaborate on the research question and the hypothesises that flow from this. As seen in the conceptual models, this paper wants to investigate three positive relationships and if these become stronger through a third variable. Each relationship will be discussed below and elaborated on.
Structural autonomy and exploitation
The first relationship is between structural autonomy as a predictor variable and exploitation as an outcome variable. As previously argued, structural autonomy refers to the extent that freedom is given to employees regarding factors that are within the work
environment (Lumpkin, 2009). In other words, structural autonomy refers to the three facets identified by Breaugh (1985); scheduling, methods and criteria autonomy. The dependent variable exploitation is conceptualised as the refinement of existing products, resources and knowledge it is associated with incremental innovations (Sinha, 2015).
Theurer, Tumasjan and Welpe (2018) stated that autonomy as a work design feature has a positive influence on creativity and innovation at work. In this study, the focus is on autonomy as dealt with in the job design literature, specifically the work of Breaugh (1985). This is coherent with the conceptualisation of structural autonomy proposed in this paper. Theurer et al. (2018) found that all three facets had a significant impact on perceived innovative behaviour at work. The reason for this positive effect is that structural autonomy activates intrinsic motivation, which positively impacts innovation in terms of idea
generation, idea promotion and implementation (Theurer et al., 2018). The link between job autonomy and intrinsic motivation is also corroborated by Van Yperen, Wörteler and De Jonge (2015).
Moreover, Orth and Volmer (2017) find that low levels of autonomy are expected to hinder employees’ attempts to innovate. Autonomy within the scope of their study is job autonomy which can be seen as what this paper defines as structural autonomy. Higher levels of job autonomy lead employees to not only generate new ideas and tackle work-related problems but to do this through innovation (Orth & Volmer, 2017). This occurs because the execution of these novel ideas in the work environment require them to be able to break out of routines and have the freedom to decide for themselves (Orth & Vomer, 2017). What is important to note from these findings, is that structural autonomy primarily affects innovation within the scope of the current work environment. This can be categorised as exploitative innovations. We, therefore, hypothesize the following:
16
Hypothesis 1: Structural autonomy has a positive effect on the exploitative activities of employees.
The moderating role of need for autonomy
The positive relationship between structural autonomy and exploitation can be influenced by several (other) factors regarding work or personal characteristics. This study proposes that the need for autonomy employees feel strengthens the relationship between structural autonomy and exploitation. To analyse this, one first needs to examine what the need for autonomy actually is. Van Yperen et al. (2015, p.603) is helpful in this respect, when they define it as: 'individual differences in the desire individuals have to experience a sense of choice and psychological freedom, and to experience oneself as the initiator and regulator of one's actions'. Their findings concluded that a low need for autonomy among workers diluted the positive effects gained from implementing more autonomy through blended working opportunities (van Yperen et al. ,2015). Langfred and Moye (2004) support this by arguing that the need for autonomy influences employee responses to task autonomy. Again, task autonomy can be categorised under the greater concept of structural autonomy. If the need for autonomy influences the extent to which individuals are motivated by task autonomy (Langfred & Moye, 2004), this then also should influence their innovative behaviour. The reason is that intrinsic motivation is one of the reasons that structural autonomy leads to an increase in creativity and innovation (Theurer et al. 2018). Taken into consideration the literature, we hypothesize the following:
Hypothesis 2: There is an interaction effect between structural autonomy and exploitation, such that the relationship between structural autonomy and exploitation becomes stronger when need for autonomy is high.
Strategic autonomy and exploration
The second conceptual model contains the other outcome variable crucial to
ambidexterity, exploration. The model insinuates that strategic autonomy has a positive effect on exploration. Strategic autonomy as defined by (Burcharth et al., 2017) is the extent to which employees have the freedom to carry out innovative activities without approval, even if they are not closely related to the firm’s strategic scope. Exploration involves activities
17
related to the discovery of new products, resources, knowledge and opportunities, and is linked to radical innovations and learning by experimentation (Sinha, 2015).
Strategic autonomy has several positive effects on activities related to exploration. First, it can provide individuals with the freedom to pursue entrepreneurial actions (Lumpkin, 2009). Second, this type of autonomy encourages workers to participate in the creation of new ventures and commercialization (Burcharth et al., 2017). Third, it encourages to proactive behaviour and personal initiative that in turns motivates employees to take more risk (Burcharth et al. 2017). Another explanation for this increased risk-taking is that employees perceive potential returns to be higher when given more strategic autonomy (Globocnik & Salomo, 2014). Enhancing exploration as risk-taking is one of its crucial elements (Volery et al.,2013). This is corroborated by Gard et al. (2013), who argue that strategic autonomy encourages lower-level managers to engage in a higher degree of risk-taking, which is necessary for the development of new businesses.
Moreover, research shows that strategic autonomy positively influences performance in a dynamic environment as it increases openness and awareness about new business
opportunities (Gard et al., 2013). It is thus strategic autonomy that increases a firm's ability to respond to a changing environment (Gard et al., 2013). The reason for this is that strategic autonomy nurtures the development of an entrepreneurial orientation, which spurs
discoveries outside the usual scope of the business (Corsino et al. 2018). One could thus argue that for a firm to achieve entrepreneurial performance, it needs strategic autonomy since structural autonomy will not suffice (Lumpkin, 2009). Given the widely shared view in the existing literature about the positive effects of strategic autonomy on employee
exploration, the following is hypothesized:
Hypothesis 3: Strategic autonomy has a positive effect on explorative behaviour employees.
Moderating role of creative self-efficacy
The relationship between strategic autonomy and explorative behaviour amongst employees can be influenced by personal differences between employees. Creative self-efficacy (CSE) can be defined as the beliefs that one has towards one’s ability to produce creative outcomes (Tierney & Farmer, 2002). This self-efficacy has revealed to be associated with creativity among individual employees (Tierney & Farmer, 2011). This, in turn, means that when creative self-efficacy becomes greater so should incidents of creative performance.
18
Creativity among employees is a crucial part of the ability of organisations to be innovative and perform successfully in dynamic environments (Tierney & Farmer, 2011). In other words, creative self-efficacy boosts innovations that are crucial for exploration activities. Not only does CSE improve explorative activities, but it also influences the previously proposed relationship between strategic autonomy and exploration. As observed by Orth and Volmer (2017), employees who believe to be capable and creative workers obtain greater benefit from autonomy for their innovative behaviour. These employees embrace the available higher levels of strategic autonomy and translate them into new innovative outcomes (Orth & Volmer, 2017). In other words, the strength of the within-person effects of autonomy on innovation can vary between the levels of creative self-efficacy between different individuals (Orth & Volmer, 2017). Taken into consideration the positive effects of CSE on innovation, this paper therefore hypothesizes the following:
Hypothesis 4: There is an interaction effect between strategic autonomy and exploration, such that the relationship between strategic autonomy and exploration becomes stronger when creative self-efficacy is high.
Contextual ambidexterity
Lastly, the relationship that the ability of a firm to engage in exploration and
exploitation at the same time will lead to contextual ambidexterity. To more formally define contextual ambidexterity is the simultaneous pursuit of exploitative and explorative activities within a single business unit (Wang & Rafiq, 2014). As stated above, this paper suggests that the two different forms of autonomy will lead to an increase in both activities. This will subsequently mean that the ability to achieve exploitation and exploration increases within the organisations. Moreover, when autonomy in both forms is implemented throughout the entire business unit it enables the simultaneous pursuit of activities within the same unit. To conclude this section, predict the following:
Proposition 1: The presence of structural and strategic autonomy will lead to increased explorative and exploitative behaviour, which is antecedent to formation of contextual ambidexterity.
19 Methods
The design used for this study was cross-sectional. Data was collected through surveys that were sent out to start-ups by e-mail and spread on social media. The exclusion criteria for filling in the survey were that respondents had to be working for an online start-up. The cut-off point to identify start-ups was the existence of a maximum of 7 years. A total of 77 respondents filled in the survey, however, 17 had to be deleted due to large
incompletion of the survey. This led to a final sample of 60 respondents. This is lower than initially expected due to the Corona crisis which led to increased uncertainty and workload for start-ups. However, this does mean that the presented results should be interpreted with caution. From the total sample 48.3% was female and 51.7% between the ages of 18-24. This high percentage of young respondents should also be taken into consideration when analysing the results.
Sample
The sample used in this study was drawn solely from employees working at online start-ups. The focus on start-ups was chosen because ambidexterity in start-ups is often overlooked but very important. Start-ups tend to either be too focused on exploring the new that they lose the scope of exploiting or over-engaged in exploitation (Sinha,2015). However, for a start-up to successfully commercialise, they must engage exploration and exploitation activities (Sinha, 2015). As start-ups are more resource constraint compared to a mature firm and have more pressure of performing for survival and sustainable growth there is a high urge for them to balance exploration and exploitation (Sinha, 2015). Difficulties arise because as mentioned before they often already have limited resources (Volery et al., 2013). As
exploration and exploitation compete for scarce resources, organisational routines and attention, this situation of limited resources is probable to amplify the paradoxes faced (Volery et al., 2013).
Measures
For this study the six main variables: exploitation, exploration, structural autonomy, strategic autonomy, creative self-efficacy and need for autonomy were all employee-rated. The final survey consisted of 34 questions and all were recorded on a 7-point Likert scale. The concepts were measured ranging from 1 = completely disagree to 7 = completely agree.
20
Exploitation. Exploitation was measured with a 5-item scale by Wang and Rafiq
(2014). Example item is ‘Over the last three years, this business has upgraded current knowledge and skills for familiar products and technologies ‘. The scale showed sufficient reliability with Cronbach’s alpha = .923
Exploration. Exploration was also measured with an adapted 4-item scale from by
Wang and Rafiq (2014). One question from the scale was omitted because of the focus of this research on start-ups. An example item was ‘Over the last three years, this business has strengthened innovation skills in areas where it had no prior experience’. The scale showed sufficient reliability with Cronbach’s alpha = .839
Structural autonomy. Structural autonomy was measured on the 9-item scale of
Breaugh (1985). For the purpose of this study only 6-items were used. Example items are ‘I am able to choose the way to go about my job (the procedures to utilize)’ and ‘My job is such that I can decide when to do particular work activities’. The scale showed sufficient
reliability with Cronbach’s alpha = .938
Strategic autonomy. Strategic autonomy was measured on a 4-item scale by
Lumpkin, Cogliser and Schneider 2009). Example item is ‘My firm, Supports the efforts of individuals and/or teams that work autonomously - Requires
individuals or teams to rely on senior managers to guide their work.’. The scale showed sufficient reliability with Cronbach’s alpha = .844
Need for autonomy. Need for autonomy was measured on a 4-item scale by Van
Yperen, Rietzschel, and De Jonge (2014). Example items are ‘I have the need to decide on my own how to go about getting my job done’ and ‘I have the need to determine on my own how to best approach my work’. The scale showed sufficient reliability with Cronbach’s alpha = .963
Creative self-efficacy. Creative self-efficacy was measured on a 3-item scale by
Tierney and Farmer (2002). Example items are ‘I feel I am good at generating novel ideas’ and ‘I can meet most creative challenges at work’. The scale showed sufficient reliability with Cronbach’s alpha = .920
21
Control variables. To rule out other possible effects on the hypothesis, control
variables have been taken into account for this research. The control variables in this study were gender, age and industry.
Analytical Plan
To test hypothesis 1, the relationship between structural autonomy and exploitation within the start-up, linear regression will be used with exploitation as the dependent variable and structural autonomy as the independent variable. To perform this regression first the assumptions needed to be tested and were all met. All variables had a VIF value lower than 5 and Tolerance value smaller dan 0.1 indicating that no multicollinearity was present.
Moreover, homoscedastic and normality test were performed, the result can be found in the appendix. To test Hypothesis 2, whether the relationship between structural autonomy and exploitation is moderated by the need for autonomy, the PROCESS macro of Hayes (2020) Model 1 will be used, with structural autonomy as the independent variable, need for autonomy the moderating variable, and exploitation as the dependent variable.
To test hypothesis 3, the relationship between strategic autonomy and exploration within the start-up, linear regression will be used with exploration as the dependent variable and strategic autonomy as the independent variable. To perform this regression same
assumptions as before needed to be tested. The output presented in the appendix shows that all assumptions were met. To test Hypothesis 4, whether the relationship between strategic autonomy and exploration is moderated by the creative self-efficacy, the PROCESS macro of Hayes (2020) Model 1 will be used, with strategic autonomy as the independent variable, creative self-efficacy the moderating variable, and exploration as the dependent variable.
22
Results
Means, standard deviations and correlations are presented below in table 1. We see a couple of important correlations in the table. The first correlation between structural
autonomy and exploitation, which is .541 with a significant p-value. This is relevant as it shows a correlation for the main relationship proposed in the first hypothesis. Related to this, the second correlation is that between the need for autonomy and exploitation, which is .589 and a significant p-value. This relates to the moderation effect proposed in hypothesis 2, the correlation between the two shows a good basis for proving it. For the third and fourth and third hypothesis, we also see that there is a significant correlation between the variables. Between strategic autonomy and exploration, there is a significant correlation of .452 and between creative self-efficacy and exploration a correlation of .439 with a significant p-value. Some correlations are interesting as they were not hypothesized. Both forms of autonomy, strategic and structural have a significant correlation for both exploitation and exploration (.463, .495). This could imply that these forms of autonomy have an effect not only on either exploration or exploitation as hypothesized but on both. This could be interesting as this paper tries to find a new pathway to contextual ambidexterity which requires both activities to be performed.
23
Table 1. Means, Standard Deviations, and Correlations
N=60 *p < .05, **p < .01
Results linear regressions
To test hypothesis 1 and 3 (there is a positive relationship between structural
autonomy and exploitation, strategic autonomy and exploration), a linear regression was used within step 1 (controls) and in step 2 (independent variable). The results showed that there was a significant relationship between independent variable structural autonomy and the dependent variable exploitation. This significant relationship was also found between the independent variable strategic autonomy and dependent variable exploration. Therefore, both H1 and H3 can be accepted.
This can be seen through several outcomes acquired in SPSS. To start with H1, the R Square of 32.8% this is the percentage of variance explained by the model. Besides, model 2 shows an R Square change of 28.6% with a p-value of .000. This indicates that the model when controlled for variables age, gender and industry explains 28.6% more variance with a significant p-value lower than .01. Lastly, the of structural autonomy, the predictor
Variables M SD 1. 2. 3. 4. 5. 6. 7. 8. 1. Age 12.84 1.309 - 2. Gender 0.50 .504 (.146) - 3. Structural autonomy 3.71 1.759 (.208) .138 - 4. Strategic autonomy 3.76 .900 .135 .001 .351** - 5. Exploitation 3.70 1.487 .001 .202 .541** .495** - 6. Exploration 3.57 1.559 .157 .092 .463** .452** .451** - 7. Need for autonomy 3.65 1.956 (.265) * .109 .871** .282* .589** .406** - 8. Creative-self-efficacy 3.11 1.737 (.225) .050 .663** .434** .528** .439** .717** -
24
variable is .457 with a t-value of 4.895. Implying that with every unit of increase of the independent outcome (structural autonomy) there is an increase of the dependent variable (exploitation) of 1.457. This increase is significant as the p-value is lower than .01.
Furthermore, we also confirmed that H3 could be accepted. This can also be seen through the outcomes acquired in SPSS. The R Square, the percentage of variance explained was 24%. The R Square change was 20.1 % with a p-value of .001. This indicates that when controlled for the variables mentioned before 20.1% variance is explained with a significant p-value of <.05. Lastly the of strategic autonomy, the predictor variable is .782 with a t-value of 3.857. This shows that with every unit increase of the independent variable (strategic autonomy) there is an increase of the dependent variable (exploration) of 1.782. This is a significant growth as it has a p-value lower than .01.
Results moderations
To test hypothesis 2 (There is an interaction effect between structural autonomy and exploitation, such that the relationship between structural autonomy and exploitation becomes stronger when the need for autonomy is present within employees), I used for PROCESS macro (Model 1) of Hayes (2020). The same was done to test hypothesis 4 (There is an interaction effect between strategic autonomy and exploration, such that the relationship between strategic autonomy and exploration becomes stronger when creative self-efficacy is present within employees). Results as shown in table 2 and 3 did not support both of the hypothesis, as they showed that there was no significant interaction effect (b=.04, se = .057, p=.488) (b= .12, se = 108, p=.264).
25 Table 2. Linear model of predictors of exploitation
β SE β t p Constant 2.477 [.89,4.06] .79 3.128 <.005 Structural autonomy -.025 [-.54,.48] .25 -.097 .923 Need for autonomy .182 [-.46, .82] .32 .569 .571 Structural autonomy x Need for autonomy .040 [-.07, .15] .06 .6968 .488 Note. R2= .36, F= 10.291, F Change = .4855
26 Table 3. Linear model of predictors of exploration
β SE β t p Constant 2.173 [-.944, 5.291] 1.556 1.396 .1681 Strategic autonomy .139 [-.722, .999] .430 .322 .7483 Creative self-efficacy -.203 [-1.073 .666] .434 -.4686 .6411 Strategic autonomy x Creative self-efficacy 1.122 [-.095, 339] .108 1.126 .2647 Note. R2= .36, F= 7.743, F Change =1.270
27 Discussion
This study examined the role of autonomy on contextual ambidexterity. More specifically, this research aimed to test whether there was a relationship between structural and strategic autonomy and exploitative and explorative activities and if this, in turn, was strengthened by the need for autonomy and creative self-efficacy. The first hypothesis stated that there was a positive relationship between structural autonomy and exploitation. The outcome was as expected: the hypothesis was supported, implying that giving employees more structural autonomy increases exploitative behaviour. The second hypothesis stated that this relationship would be strengthened when employees felt a high need for autonomy. In this case, results did not lend significant evidence for this. For the second model, the third hypothesis stated that higher levels of strategic autonomy would positively impact
exploration within the firm and was supported. However, the fourth hypothesis that proposed a stronger relation between strategic autonomy and exploration when there were high levels of creative self-efficacy was not significant.
The outcome that structural autonomy positively impacts exploitation is confirms the results of earlier studies. Orth and Volmer (2017) e.g. concluded that lower levels of
autonomy hindered employees from innovation. In addition, Theurer et al. (2018)
corroborated this by concluding that all three facets from Breaugh (1985) had an impact on perceived innovative behaviour. More specifically, Corsino et al. (2018) validated the view that structural autonomy led to innovation within the scope of the business.
The findings in this paper are not only in line with these previous findings but are able to add significant value. In current the literature, innovative behaviour has not been linked to exploitation within the ambidexterity framework. From the description of what is viewed as innovative behaviour, such as Orth & Volmer (2017), who specify it as generating novel ideas for work-related problems, one can see that there is a link to exploitation. As mentioned before, exploitation concerns incremental innovations that are within the current scope of the business (Wang & Rafiq, 2014). The results of this study put the previously discovered and reaffirmed relationship into the broader context of the debate on
ambidexterity. This is interesting as it enables scholars to discern a connection between the broad field of autonomy and the increasingly important field of ambidexterity.
This link can also be seen in the confirmation of the third hypothesis of the paper that stated that strategic autonomy would positively influence explorative behaviour amongst employees. When looking back at the theoretical framework, we see that both Burcharth et al. (2017) and Gard et al. (2013) published results in line with the ones found in this research.
28
The former et al. (2017) argued that strategic autonomy encouraged workers to participate in the creation of new ventures, while the latter concluded that it would increase the firm’s ability to respond to changes characteristic of volatile markets. Again, we see a clear link to what can be described as explorative behaviour, i.e. a firm’s ability to move towards new opportunities and adjust to volatile markets (Birkinshaw and Gibson, 2004).
This connection between the findings of this paper and the broader concept of contextual ambidexterity can be viewed as support for proposition 1, mentioned in the theoretical framework. This argued that the increase in exploration and exploitation would accordingly mean an increase in the formation of contextual ambidexterity. A key aspect of contextual ambidexterity is the simultaneous pursuit of exploitation and exploration
(Birkinshaw & Gibson, 2004). It enables individuals to make their own choices regarding the balance between exploration and exploitation in their work (O'Reilly III & Tushman, 2013). The aforementioned results have indicated that an increase in both forms of autonomy leads to an increase in exploration and exploitation. The essence of contextual ambidexterity is the dual capacity in these activities; thus, one could argue that strategic and structural autonomy indirectly increase the formation of contextual ambidexterity.
This study did not find support for the claim that the need for autonomy strengthens the relationship between structural autonomy and exploitation. This indicates that, in contrary to the findings by Van Yperen et al. (2016) and Langfred and Moye (2004), the need for autonomy among workers does not shape the effects of structural autonomy on exploitation. This is interesting as it implies that the positive effect on exploration would not have to be accompanied by a high need for autonomy amongst employees. Langfred and Moye (2004) suggested that the positive effect of structural autonomy on innovation would be diluted when need for autonomy was not present, this study contradicts these findings. This tells us that the supposed beneficial effect that companies strive for can be achieved regardless of the need for autonomy felt by its employees.
In addition, the results for the fourth hypothesis did not lend substantial evidence to be confirmed. Implying that a high level of creative self-efficacy does not lead to more explorative behaviour when strategic autonomy is present. This is also of interest as it contrasts the findings by Orth and Volmer (2017), who propositioned that the effects of autonomy on innovation could vary between the level of CSE among employees. Again, we see a positive outcome to this non-significance as it implies that exploration will increase independent of how creative employees perceive themselves.
29
An alternative explanation for this discrepancy with the literature may be related to the differences in the research population. Previous studies such from Orth and Volmer (2017) and Van Yperen et al. (2016) worked with a wide variety of industries and firm sizes, whereas this project focused on online start-ups only. It is not unlikely to expect that the level of creative self-efficacy and need for autonomy amongst employees of start-ups will differ from those working for larger firms in view of the cultural differences between these types of firms.
Limitations and further research
It is important to take into consideration that there could also be other explanations for the above-acquired results. First, contextual ambidexterity and the related activities exploration and exploitation are influenced by several characteristics of culture, values and management practices. It could be that contextual ambidexterity indeed increases with structural and strategic autonomy but that this is only a small portion compared to other firm characteristics. However, the study presents important findings as it demonstrates that there was a significant relationship. Start-ups struggling to find pathways to contextual
ambidexterity could increase their levels of autonomy to facilitate the process. However, they need to take into consideration that achieving ambidexterity requires many more changes.
When analysing this study some limitations need to be discussed. The first concerns the limited amount of surveys that could be collected. As mentioned before, the impact of the Covid-19 on start-ups was and continues to be has been significant and has increased work pressure greatly. It was understandably very hard to contact employees willing to participate in the survey. As a consequence, the results should be interpreted with care. Secondly, contextual ambidexterity was not measured separately but assumed as exploitation and exploration both increased. Even though contextual ambidexterity is the presence of both of these activities, this should be taken into consideration
In view of the limitations and outcome of this study several implications for further research can be formulated. The first builds upon the weakness of the sample. The research could be repeated using a much larger sample, which would increase the generalisability of the findings. Second, this paper solely focused on the relationship between autonomy and ambidexterity in online start-ups. It is interesting to investigate whether this relationship would also hold for medium-sized to large organisations. Third, as this research did not explain what could strengthen or diminish the relationship between autonomy and contextual ambidexterity, further research needs to be done. Lastly, contextual ambidexterity could be
30
measured as a separate construct to test if the proposed relationship between structural and strategic autonomy and contextual ambidexterity holds.
Managerial implications
As mentioned, the most important goal of this research was to find a new approach to facilitate contextual ambidexterity. Considering the result, we saw that strategic and
structural autonomy had a positive impact on exploration and exploitation. This meant that the presence of both led to the formation of contextual ambidexterity. This implies a new route to contextual ambidexterity, which is the foundation for practical implications.
Managers seeking to increase the simultaneous pursuit of exploration and exploitation need to take a close(r) look at the level of autonomy in the firm. It is worth considering granting workers autonomy on a structural level, such as enabling them to have freedom in choosing what methods to use in their day to day work (Breaugh, 1985)., as well as on a strategic level with regards to goal setting (Lumpkin, 2009). The presence of both will facilitate the process of turning the start-up into an ambidextrous one.
Conclusion
This research project builds on and contributes to previous scholarly research about the different ways in which ambidexterity in organisations and companies can be structured. Interestingly, concrete management practices that facilitate strengthening ambidexterity have so far not received ample attention. In a time of marked volatility and uncertainty such a research topic seemed particularly appropriate and relevant. Therefore, the present study set out to study the effects of structural and strategic autonomy on explorative and exploitative activities in firms, the underlying question being that they would lead to the increased formation of contextual ambidexterity. The research population consisted of start-up
companies in the Netherlands. Moreover, this paper wanted to investigate if this relationship would be strengthened when the need for autonomy and creative self-efficacy amongst employees was high. The results show that both forms of autonomy indeed have a positive influence on exploitation and exploration and thus on contextual ambidexterity within start-ups. The study also found that ambidexterity was not strengthened by the need for autonomy or creative self-efficacy. This outcome has two practical implications. First, when start-ups are interested in finding new pathways to achieve contextual ambidexterity it would be beneficial to provide for an environment with higher levels of autonomy on both the strategic and structural level. Second, this beneficial effect occurs despite the level of need for
31 Reference list
Breaugh, J. A. (1985). The Measurement of Work Autonomy. Human Relations, 38(6), 551– 570. https://doi.org/10.1177/001872678503800604
Breaugh, J. A. (1999). Further Investigation of the Work Autonomy Scales: Two Studies. Journal of Business and Psychology, 13(3), 357–373.
https://doi.org/10.1023/a:1022926416628
Birkinshaw, Julian & Gibson, Cristina. (2004). Building Ambidexterity into an Organization. MIT Sloan Management Review. 45.
Burcharth, A., Præst Knudsen, M., & Søndergaard, H. A. (2017). The role of employee autonomy for open innovation performance. Business Process Management Journal, 23(6), 1245–1269. https://doi.org/10.1108/bpmj-10-2016-0209
Chen, Y. (2017). Dynamic ambidexterity: How innovators manage exploration and exploitation. Business Horizons, 60(3), 385–394.
https://doi.org/10.1016/j.bushor.2017.01.001
Corsino, M., Giuri, P., & Torrisi, S. (2018). Technology spin-offs: teamwork, autonomy, and the exploitation of business opportunities. The Journal of Technology Transfer, 44(5), 1603–1637. https://doi.org/10.1007/s10961-018-9669-1
Gibson, C. B., & Birkinshaw, J. (2004). THE ANTECEDENTS, CONSEQUENCES, AND MEDIATING ROLE OF ORGANIZATIONAL AMBIDEXTERITY. Academy of Management Journal, 47(2), 209–226. https://doi.org/10.2307/20159573
32
He, Z.-L., & Wong, P.-K. (2004). Exploration vs. Exploitation: An Empirical Test of the Ambidexterity Hypothesis. Organization Science, 15(4), 481–494.
https://doi.org/10.1287/orsc.1040.0078
Jansen, J. J. P., Tempelaar, M. P., van den Bosch, F. A. J., & Volberda, H. W. (2009). Structural Differentiation and Ambidexterity: The Mediating Role of Integration Mechanisms. Organization Science, 20(4), 797–811.
https://doi.org/10.1287/orsc.1080.0415
Gard, J. , G. Baltes, Wehle D., & Katzy D., "An integrating model of autonomy in corporate entrepreneurship," 2013 International Conference on Engineering, Technology and Innovation (ICE) & IEEE International Technology Management Conference, The Hague, 2013, pp. 1-14.
doi: 10.1109/ITMC.2013.7352658
Globocnik, D., & Salomo, S. (2014). Do Formal Management Practices Impact the Emergence of Bootlegging Behavior? Journal of Product Innovation Management, 32(4), 505–521. https://doi.org/10.1111/jpim.12215
Langfred, C. W., & Moye, N. A. (2004). Effects of Task Autonomy on Performance: An Extended Model Considering Motivational, Informational, and Structural
Mechanisms. Journal of Applied Psychology, 89(6), 934–945. https://doi.org/10.1037/0021-9010.89.6.934
Lopes, H., Lagoa, S., & Calapez, T. (2014). Work autonomy, work pressure, and job satisfaction: An analysis of European Union countries. The Economic and Labour Relations Review, 25(2), 306–326. https://doi.org/10.1177/1035304614533868
33
Lumpkin, G. T., Cogliser, C. C., & Schneider, D. R. (2009). Understanding and Measuring Autonomy: An Entrepreneurial Orientation Perspective. Entrepreneurship Theory and Practice, 33(1), 47–69. https://doi.org/10.1111/j.1540-6520.2008.00280.x
O’Reilly, C. A., III, & Tushman, M. L. (2013). Organizational Ambidexterity: Past, Present, and Future. Academy of Management Perspectives, 27(4), 324–338.
https://doi.org/10.5465/amp.2013.0025
Orth, M., & Volmer, J. (2017). Daily within-person effects of job autonomy and work engagement on innovative behaviour: The cross-level moderating role of creative self-efficacy. European Journal of Work and Organizational Psychology, 26(4), 601– 612. https://doi.org/10.1080/1359432x.2017.1332042
Sadler - Smith, E., El-Kot, G., & Leat, M. (2003). Differentiating work autonomy facets in a non-Western context. Journal of Organizational Behavior, 24(6), 709–731.
https://doi.org/10.1002/job.20
Sinha, S. (2015). The Exploration–Exploitation Dilemma: A Review in the Context of Managing Growth of New Ventures. Vikalpa: The Journal for Decision Makers, 40(3), 313–323. https://doi.org/10.1177/0256090915599709
Theurer, C. P., Tumasjan, A., & Welpe, I. M. (2018). Contextual work design and employee innovative work behavior: When does autonomy matter? PLOS ONE, 13(10),
e0204089. https://doi.org/10.1371/journal.pone.0204089
Tierney, P., & Farmer, S. M. (2002). CREATIVE SELF-EFFICACY: ITS POTENTIAL ANTECEDENTS AND RELATIONSHIP TO CREATIVE PERFORMANCE. Academy of Management Journal, 45(6), 1137–1148.
34
Tierney, Pamela, & Farmer, S. M. (2011). Creative self-efficacy development and creative performance over time. Journal of Applied Psychology, 96(2), 277–293.
https://doi.org/10.1037/a0020952
Van Yperen, N. W., Rietzschel, E. F., & De Jonge, K. M. M. (2014). Blended Working: For Whom It May (Not) Work. PLoS ONE, 9(7), e102921.
https://doi.org/10.1371/journal.pone.0102921
Van Yperen, N. W., Wörtler, B., & De Jonge, K. M. M. (2016). Workers’ intrinsic work motivation when job demands are high: The role of need for autonomy and perceived opportunity for blended working. Computers in Human Behavior, 60, 179–184. https://doi.org/10.1016/j.chb.2016.02.068
Volery, T., Mueller, S., & von Siemens, B. (2013). Entrepreneur ambidexterity: A study of entrepreneur behaviours and competencies in growth-oriented small and medium-sized enterprises. International Small Business Journal: Researching
Entrepreneurship, 33(2), 109–129. https://doi.org/10.1177/0266242613484777
Wang, C. L., & Rafiq, M. (2014). Ambidextrous Organizational Culture, Contextual Ambidexterity and New Product Innovation: A Comparative Study of UK and Chinese High-tech Firms. British Journal of Management, 25(1), 58–76. https://doi.org/10.1111/j.1467-8551.2012.00832.x
35 Appendix
Reliability Exploitation
36 Reliability Structural autonomy
37 Reliability Need autonomy
Reliability Creative self-efficacy
38
Correlations: age, gender, industry, structural autonomy, strategic autonomy, need for autonomy, creative self-efficacy, exploitation and exploration
Regression assumptions
Normality- Structural autonomy against exploitation
39 Normality – Need for autonomy against exploitation
40 Normality – Structural autonomy against exploration
41 Normality – Creative self-efficacy against exploration
42 Multicollinearity: Structural autonomy
43 Multicollinearity: Need for autonomy
Multicollinearity: Strategic autonomy
Multicollinearity: Exploration
44 Homoscedasticity: Structural autonomy against exploitation
45 Homoscedasticity: Strategic autonomy against exploration
46
Linear regression structural autonomy and exploitation without control variables
Linear regression structural autonomy and exploitation including control variables
Linear regression strategic autonomy and exploration without control variables
47 Process model 1 Hayes, need for autonomy as moderator
