Maria Pajuoja From mechanistic measuring to up-to-date understanding Problematising the study of innovative work behaviour  ACTA WASAENSIA 489 Copyright © University of Vaasa and the copyright holders. ISBN 978-952-395-028-3 (print) 978-952-395-029-0 (online) ISSN 0355-2667 (Acta Wasaensia 489, print) 2323-9123 (Acta Wasaensia 489, online) URN http://urn.fi/URN:ISBN:978-952-395-029-0 Hansaprint Oy, Turenki, 2022. ACADEMIC DISSERTATION To be presented, with the permission of the Board of the School of Management of the University of Vaasa, for public examination in 2nd of September, 2022, at noon. Article based dissertation, School of Management, Human Resource Management. Author Maria Pajuoja Supervisor(s) Professor Riitta Viitala University of Vaasa, School of Management, Human Resource Management. DSc. (econ.) Susanna Kultalahti University of Vaasa, School of Management, Human Resource Management. Custos Professor Riitta Viitala University of Vaasa, School of Management, Human Resource Management. Reviewers Professor Joakim Wincent Hanken School of Economics, Department of Entrepreneurship, Management and Organisation. Associate Professor Timothy Bednall Swinburne University of Technology, Department of Management and Marketing. Opponent Associate Professor Timothy Bednall Swinburne University of Technology, Department of Management and Marketing. V Tiivistelmä Innovaatiot ovat elintärkeitä yritysten selviytymiselle. Yksittäiset työntekijät, jotka luovat ja implementoivat uusia ideoita, ovat organisaatiotason innovaatioiden ja innovatiivisuuden perusta. Yksilötason innovaatioprosesseja tutkitaan usein termillä innovatiivinen työkäyttäytyminen, joka on käsillä olevan tutkimuksen lähtökohtana. Hermeneuttisessa tutkimuksessa etsitään nyt vastauksia siihen, voiko valmentava esihenkilötyö lisätä yksilön innovatiivista työkäyttäytymistä, miten tämä työkäyttäytyminen on ymmärretty ja miten sitä mitataan, ja onko yksilötason innovaatioprosessi samanlainen kuin aikaisempi tutkimus on sen käsittänyt. Väitöskirjan tavoitteena on lisätä ymmärrystä yksilön innovatiivi- suudesta ja siitä, miten sitä tulisi lähestyä tutkimuksessa ja työelämässä. Ilmiötä lähestytään monimenetelmällisesti yhdistäen määrällisiä ja laadullisia menetelmiä ja erilaisia aineistoja. Kvantitatiivinen aineisto koostuu 4418 pk- sektorin työntekijöiden vastauksesta. Kirjallisuuskatsauksessa analysoidaan 255 artikkelin aineistoa. Tapaustutkimus pohjautuu 34 puolistrukturoituun teema- haastatteluun. Tutkimus osoittaa, että yksilön innovaatioprosessien tutkiminen termillä innovatiivinen työkäyttäytyminen on ongelmallista ja selittää syitä tähän. Yksilön innovaatioprosessia on viimeksi tutkittu 1980-luvulla, ja näihin tutkimuksiin perustuvat vielä nykyäänkin käytössä olevat mittaristot. Väitöskirjassa esitetään, että yksilön innovaatioprosessi on nykyään erilainen kuin 1980-luvulla, mm. siihen kuuluvien aktiviteettien ja yksilön roolin osalta. Innovatiivisen työ- käyttäytymisen tutkimuksessa on keskitytty mekanistiseen käyttäytymiseen vaikuttavien tekijöiden mittaamiseen vanhentuneilla mittaristoilla sen sijaan, että olisi yritetty ymmärtää sitä, miten yksilöt nykyään innovoivat. Väitöskirja rakentaa uudenlaista kuvaa yksilötason innovaatioprosessista lisäten ymmärrystä siitä sekä tarjoaa kiintopisteitä yksilön innovatiivisuuden tukemiseen organisaatioissa. Asiasanat: Innovatiivinen työkäyttäytyminen, yksilötason innovaatioprosessi, valmentava esihenkilötyö, hermeneuttinen kehä, monimenetelmällisyys VI Abstract Innovation is vital for the survival of organisations. Individual employees are the microfoundations of organisational innovation since it is the individuals who generate new ideas and implement solutions. Individual innovation processes are often studied using the term innovative work behaviour (IWB). The topic is explored in a hermeneutic circle in which answers are sought to whether managerial coaching can enhance IWB; how IWB has been understood and measured; and whether the individual innovation process is similar to how previous literature has conceived it. The overall aim of the thesis is to increase understanding of the individual innovation process and how it should be approached in research and in modern working life. Three data sets are utilised in the quest for answers. 4418 responses from employees in Finnish SMEs comprise the quantitative data set. An article cache of 255 articles is analysed in the integrative literature review. 34 semi-structured interviews at a Finnish MNC make up a single case study. The mixed-methods approach allows for a multifaceted understanding of the phenomenon studied. The thesis makes several important contributions. It highlights that the practice of studying individual innovation processes under the term innovative work behaviour is problematic, and explains reasons for this. It finds that the individual innovation process was last studied in the 1980s and that the measuring instruments used even today are based on these studies. The thesis suggests that the individual innovation process is different today than it was in the 1980s, both in consisting of different activities than before, and in the role of the individual being more active and engaged. In all, the study of IWB has focused on the mechanistic measuring of the effects of various determinants to behaviours using outdated measuring instruments instead of attempting to understand the microfoundations of innovation. The thesis builds new understanding of the individual innovation process and offers focal points for supporting an individual’s innovation efforts in organisations. Keywords: Innovative work behaviour, individual innovation process, managerial coaching, hermeneutic circle, mixed methods VII ACKNOWLEDGEMENT This is it – I have achieved my childhood dream of becoming a published author. I now feel like I should have been more specific as a child since it appears that getting published does not automatically mean international book tours and getting to decide whether Anne Hathaway or Natalie Portman should play me in the movie. That may be just as well since, in the end, watching a movie about a 40- something-year-old woman spending most of her day with her head resting on the desk, sighing in desperation, would not make for very good entertaining, no matter who she is portrayed by. Another thing that I did not realise is that, much like with raising kids, it takes a village. Fortunately, I had a very supportive village around me. My supervisor, boss, mentor, and friend, professor Riitta Viitala was an invaluable companion, alternately pushing me to produce better-quality results and reminding me to cut myself some slack. I could fill up a couple of pages just listing her many merits but I will just say this: the fact that she never saw me as anything other than an equal ended up being perhaps the characteristic that meant the most to me. My second supervisor, assistant professor Susanna Kultalahti, could always be counted on for top-notch advice on my work, funding applications, career, and how to throw a good party. I also want to mention assistant professor Anni Rajala for answering literally hundreds of (mostly stupid) questions about quantitative research. Professor Adam Smale, in the role of our school’s dean, hired me early on in the process, thus providing me with time and motivation to focus on my research. The pre-examination process was very helpful in giving me a glimpse of how experts in my field read my work. I am grateful to professor Joakim Wincent and associate professor Timothy Bednall for their insightful comments and suggestions which gave me a chance to improve the manuscript. I want to thank Dr Timothy Bednall for accepting the invitation to act as the opponent in the examination; I look forward to a fruitful and interesting discussion. My colleagues have been invaluable throughout this process. In particular, I want to call out my co-authors Maarit Laiho and Kaisa Henttonen for teaching me so much about the research process and how it leads to a written article; Marko Kohtamäki, Rumy Narayan, and Rodrigo Rabetino for discussions on the philosophy of science and the nature of knowledge; and Kati Söderlund, Suvi Einola, Niina Koivunen, Jukka Vesalainen, Tiina Jokinen, and Maria Järlström for well-placed pieces of advice and kind words of encouragement. For peer support, pick-me-ups, and information sharing, I am grateful to my fellow PhD candidates VIII Tania Biswas, Rodrigo Mello, Inés Escobar Borruel, Laura Urrila, and Elizabeth Edgal, with special thanks to Tiina Leino for encouraging me to go for a PhD in the first place. Thank you, Heini Pensar, for the walks, talks, lunches, and thought- provoking questions, and Jenni Kantola for challenging my thinking and for being a great companion in Barcelona. Beyond my home university, I am happy to have found a supportive and fun research community at ISPIM; the presentations and comments at conferences have been extremely useful, and interactions with like- minded people have resulted in some of the most amazing moments on this journey (I’m thinking learning line dancing on a Thai riverboat). The generous funding that I have been able to obtain has made living in this village highly enjoyable, and I wish to acknowledge the Foundation for Economic Education (Liikesivistysrahasto), and the Evald and Hilda Nissi Foundation for the support they have given me. The OpenInnoTrain project, funded by European Union’s Horizon 2020 research and innovation programme under the Marie Sklodowska-Curie grant, enabled me to go on research visits to Melbourne and Barcelona and led me to get to know such inspiring academics as professor Anne- Laure Mention and warm-hearted, open-minded, and hilarious people as Bruno Woeran and Georg Macher. Finally, the support on the home front has been my lifeline. I have been blessed with two amazing friends and fellow writers: Eva-Maria Oberauer comforted me during those times when I wanted to pluck out my eyebrows one by one, and Johanna Virtanen had such amazing faith in me that it was impossible not to have faith in myself. My parents and siblings with their multitude of kids make up an Italian-style, loud and chaotic family life that offers a refreshing change from the (occasional) solemnity of the academic life. Anna-Lena Walleczek made everything easier by competently and with exceptional judgment managing our household during the last months leading to the completion of this thesis. My helpful, wise, and innovative husband Sami has played all the roles you can think of: housekeeper, chef, massage therapist, IT support, cheerleader, and discussant, to name but a few. He is the best companion a girl could have asked for in this – or, indeed, any – endeavour. Noa and Amos have been responsible for ensuring that I take my mind off work by demanding that I make them pancakes, race toy cars, and read them stories about bears who fart. Guess what, boys: mummy is finally done with her big book. Barcelona, June 2022 Maria Pajuoja IX Contents ACKNOWLEDGEMENT ............................................................................ VII 1 PROLOGUE ........................................................................................ 1 2 INTRODUCTION ................................................................................. 3 2.1 The importance of innovation .................................................. 3 2.2 Academic discussions on innovation ........................................ 5 2.3 Defining innovative work behaviour ......................................... 7 2.4 Problematising the study of innovative work behaviour .......... 10 2.4.1 Current understanding of innovative work behaviour hails from the 1980s ............................................... 11 2.4.2 Behaviour is not the same as process ...................... 13 2.4.3 The study of innovative work behaviour is confused about the correct level of analysis ........................... 14 2.5 Objective and research questions .......................................... 16 2.6 Thesis structure ..................................................................... 18 3 THEORETICAL BACKGROUND ........................................................... 19 3.1 The study of innovation ......................................................... 19 3.1.1 Innovation outcomes ............................................... 20 3.1.2 Innovation processes .............................................. 21 3.1.3 Determinants of innovation ..................................... 23 3.2 The individual innovation process .......................................... 24 3.3 Measuring innovative work behaviour .................................... 27 3.4 Managing innovative work behaviour ..................................... 29 3.4.1 Leadership styles .................................................... 30 3.4.2 Supervisory practices .............................................. 31 3.4.3 Issues with managerial determinants ...................... 32 3.4.4 Managerial coaching as a determinant of innovative work behaviour ....................................................... 33 3.5 The state-of-science of the concept of innovative work behaviour .............................................................................. 34 4 RESEARCH METHODOLOGY .............................................................. 36 4.1 Methodological choices ......................................................... 36 4.2 Research methods ................................................................. 38 4.3 Data sets and analyses .......................................................... 40 5 SUMMARY OF STUDIES ..................................................................... 44 5.1 Managerial coaching affects different dimensions of IWB differently .............................................................................. 45 5.2 Managers influence their employees’ IWB through work engagement .......................................................................... 46 5.3 Current measuring instruments may not reflect the IWB of today ..................................................................................... 48 5.4 The individual innovation process looks different than previously thought ................................................................ 50 X 5.5 Summary of research results ................................................. 52 5.6 Quality assessment of the studies.......................................... 53 5.6.1 Evaluating the quantitative studies .......................... 53 5.6.2 Evaluating the qualitative studies ............................ 54 6 DISCUSSION AND OVERALL CONTRIBUTIONS ................................... 56 6.1 Theoretical contributions ....................................................... 56 6.2 The future of innovative work behaviour ................................ 60 6.3 Methodological contributions ................................................ 64 6.4 Practical implications ............................................................. 65 6.5 Limitations ............................................................................ 67 6.6 Reflections ............................................................................ 68 REFERENCES ......................................................................................... 70 APPENDICES .......................................................................................... 85 Appendix 1. Measurement scales .................................................... 85 Appendix 2. Interview guide ............................................................ 87 PUBLICATIONS ...................................................................................... 88 Figures Figure 1 Levels and streams of innovation research ....................... 5 Figure 2. Nested nature of questions about IWB .............................. 8 Figure 3. Initial research interest ................................................... 20 Figure 4. Continuum of basic assumptions in social science research ......................................................................... 36 Figure 5. Points of view of the research papers ............................. 44 Figure 6. Potential directions for further studies ........................... 63 Figure 7. Different roles in the innovation process ........................ 66 Tables Table 1. Description of the hermeneutic research process ........... 17 Table 2. Previously suggested dimensions of IWB ........................ 26 Table 3. Data sets ....................................................................... 40 Table 4. Summary of the contributions ........................................ 53 XI Abbreviations HRM Human resource management IWB Innovative work behaviour LMX Leader-member exchange theory MC Managerial coaching R&D Research and development SME Small- and medium-sized enterprises WE Work engagement XII Publications [1] Pajuoja, M., R. Viitala & K. Henttonen (in progress). Examining the effect of managerial coaching on four dimensions of innovative work behaviour. An earlier version of the paper was presented at the ISPIM conference in Bangkok, Thailand in March 2020. [2] Laiho, M., R. Viitala, M. Pajuoja & K. Henttonen (in progress). Managerial coaching and employees’ innovative work behaviour – The mediating effect of work engagement. [3] Pajuoja, M. (in progress). Are we measuring the innovative work behaviour of the 1980s? A critical review of the measuring instruments. Paper presented at the virtual EURAM conference in June 2021. [4] Pajuoja M. & M. Laiho (in progress). Insights into the individual innovation process: Updating prevalent understanding. Paper accepted for presentation at the Academy of Management conference in Seattle in August 2022. 1 PROLOGUE “Get mad about what’s missing in your conversation to get over the timidity newcomers often feel.” Anne Sigismund Huff, former President of the Academy of Management I am eager for us to get started on a topic that I find to be of the utmost importance: what individuals do when they innovate. Although innovation is usually a team effort, individuals still act in certain ways as part of the team, and understanding how their processes unfold is, for me, extremely interesting. Indeed, as a former manager at one of the most innovative companies in the world, I witnessed daily how my colleagues innovated and how the results of their processes allowed us to do our jobs faster, more effectively, and have more fun while doing it. Naturally, understanding an individual’s role in the innovation process was always going to be my research topic. But before delving into that, I want to spend a couple of minutes explaining two transitions that occurred during my thesis journey. They help explain why I chose this topic, and why the thesis took the direction that it did. From linear progression to hermeneutic circle I started with the question that seemed highly practical to me: what can managers do to help their employees be more innovative? I explored this topic in the first two papers, expecting to find answers straightforward enough to follow in a neat, linear fashion. I am usually not happy to just accept answers that are given to me, however. So, in a way, it came as no surprise that when I started to conduct research into how managers can influence the innovative work behaviour of their employees, my natural instinct was first to question whether the way innovative work behaviour was measured is correct; then, whether the whole concept in itself is accurate and up-to-date; and ultimately whether the question of how managers can “influence” their employees is the most pertinent one. Eventually, I came to understand that I was engaged in a hermeneutic process of interpreting and understanding the concepts and data before me. I particularly relate to Gadamer’s definition of the hermeneutic circle which emphasises the 2 Acta Wasaensia development of a new understanding of a concept after exploring its details in an iterative process that goes back and forth between the whole and its parts (Gadamer, 2004). Hermeneutic understanding arises in a dialogue between the interpreter and what is to be understood with the interpreter’s prior experiences, knowledge, and prejudices as essential elements in the quest to build new understanding (Gadamer, 2004). The use of the word circle has been criticised as incorrect; for who or what exactly is at the centre of the circle (Shklar, 2004)? I experienced the circle more as a spiral but will use the word circle as it is the established term. From positivist to more interpretative ponderings I was all set to be a positivist. Having finally mastered enough matrix algebra to do quantitative data analysis I took to it like fish to water and revelled at how beautifully the numbers aligned and painted such compelling pictures. When I realised that I would not be able to find out what I wanted to find out for Paper 4 utilising quantitative research methods I grudgingly enrolled on a post-graduate course on qualitative methods. “I live my life thinking of everything I see as latent variables and analysing which way the arrows might go from one variable to another. Trying now to conceptualise what a case study might look like and what I can accomplish with that is like trying to breathe under water,” I wrote in my learning diary for the course. My most significant qualms related to my own role in making science and I have frequently been reluctant to take what I considered a more prominent role than I was ready to accept. How do I safeguard science from my mistakes and deficiencies in understanding? How do I make sure that the lens through which I observe as a researcher is reliable? And ultimately, I suppose; why should anyone be interested in any of the thoughts and ideas that are the result of my subjective processing? Obviously, I am at the beginning of my quest to answer some of these questions and the struggle is, I think, evident in these pages. But, to go back to the quote that started this prologue, I did get mad about what I felt was missing in the conversation about how individuals innovate. And that did help me get over some of my feelings of being a newcomer to research. Acta Wasaensia 3 2 INTRODUCTION This chapter starts with an exploration of the importance of innovation and what, and how, has been studied about innovation. I have here considered the general, vast discussion on innovation to give the reader a chance to place my research on the field. I will then narrow down the focus to my interest area, how individuals innovate. This allows for a more detailed discussion of what that study looks like, what some of the issues with it are, and what the objective of this thesis is. 2.1 The importance of innovation The World Economic Forum (Strategic Intelligence, n.d). lists a vast number of global issues to be resolved. Although the issues are complex, and as such, resist any attempts at categorising them, broad distinctions can be made between technological (such as Artificial Intelligence and the Internet of Things), ecological (e.g. climate change and biodiversity), economical (e.g. digital economy, circular economy), societal (e.g. ageing and systemic racism) and governance issues (e.g. internet governance, global governance). That we are facing unprecedented changes is clear, and the speed of technological disruption makes the situation even more complex. The OECD has called for a re-setting of policies to ensure that innovation efforts are directed towards resolving the issues that we are faced with . A good example of innovating to fight a global issue is the currently ongoing global pandemic COVID-19. Innovation has played an essential role both in understanding what the virus is like and how it is transmitted and also in developing vaccines in a short period of time (OECD Science, Technology and Innovation Outlook - OECD, n.d). At the national level, the ability to innovate and to bring those innovations to market has long been understood to be a crucial factor in how nations remain globally competitive and many countries have produced national strategies for their innovation efforts (The OECD Innovation Strategy: Getting a Head Start on Tomorrow, 2010). Finland is no exception and a 2019 report on Finland’s innovation policy clearly nominates innovation (and specifically internationally successful innovation) as the solution to the sustainable growth of the economy and rising employment rates. (Koski, Husso, Kutinlahti, Huuskonen, & Nissinen, 2019). That innovation is important for the success and survival of organisations is so much a given today that many articles bypass the entire claim with a short single sentence at the beginning (e.g., Anderson, Potočnik, & Zhou, 2014; Hughes, Lee, Tian, Newman, & Legood, 2018), or, indeed, make no mention of it at all (e.g., 4 Acta Wasaensia Lukes & Stephan, 2017). Digging deeper into the literature reveals that there are two main areas that respond to innovation that can help organisations be more competitive: the products and services the organisation offers (product innovation) and how those offerings are delivered (process innovation) (Francis & Bessant, 2005). Product innovation refers to the ability to introduce new products or services to benefit customers, or to exploit technologies commercially (Gopalakrishnan & Damanpour, 1997; Lukas & Ferrell, 2000). Focusing on improving processes can help the organisation become more effective, improve their quality, and save costs (Damanpour, 1991; Johne, 1999). Both types of innovation can be either incremental or radical, and in their radical form can transform industries and destroy competition (Gopalakrishnan & Damanpour, 1997). Additionally, and relevant to this thesis, how innovation is managed is also seen as a way of ensuring competitive success (Adams, Bessant, & Phelps, 2006). Suggestions include developing managers’ ambidextrous capabilities to allow them to manage both radical and incremental innovation (Tushman & O’Reilly, 1996), and learning how to harness the knowledge and capabilities of employees at all levels (Dess & Picken, 2000). Regardless of whether the advantage comes from new products, improved processes, or better management of innovation, that innovation is more critical now than ever is clear from a report by McKinsey in 2020. In the survey of over 200 organisations representing various industries, 90% of the executives believed that COVID-19 would fundamentally change their business in the next five years. Over 75% thought that the crisis would create significant new growth opportunities. However, only 21% said that they had the skills and resources to pursue these opportunities. (Bar Am, Furstenthal, Jorge, & Roth, 2020). Clearly, the need to innovate and generate knowledge on innovation is as important as ever. Individuals also benefit from innovation efforts. Although research on this topic is sparse (Janssen, van de Vliert, & West, 2004), it has been shown that engaging in innovative behaviours leads to enhanced employee engagement and well-being (Huhtala & Parzefall, 2007) and can even shield employees from burnout following their company downsizing (Hammond, Cross, Farrell, & Eubanks, 2019). Engaged employees tend to innovate more (van Zyl, van Oort, Rispens, & Olckers, 2019; Wu & Wu, 2019) which makes innovation an activity that results from feeling positive, vigorous, and fulfilled at work (Schaufeli, Salanova, González-Romá, & Bakker, 2002). Therefore, the circumstances necessary to encourage an individual to innovate call for both capabilities to innovate and a highly engaged state of mind. The individual innovation process is as much a mental process as a technical one, Acta Wasaensia 5 and the manager’s role in facilitating the process has been reported to be instrumental (e.g., Hughes et al., 2018). 2.2 Academic discussions on innovation Innovation can be studied at the individual, work team, organisational, or at multiple levels (Anderson et al., 2014). Three streams of innovation research can be distinguished: one that focuses on the determinants of innovation, one that looks at innovation as a process, and one that is interested in innovation outcomes (Crossan & Apaydin, 2010). The streams are related to each other: determinants affect the innovation process, and the innovation process must precede innovation outcomes (Crossan & Apaydin, 2010). Putting these two categorisations together gives us Figure 1, which presents how the levels and streams merge. The details of Figure 1 are discussed in Chapter 3. Figure 1 Levels and streams of innovation research From the start, I was interested in how organisational-level determinants (in particular managerial actions) affect individual-level innovation processes. When talking about these individual processes, the term innovative work behaviour (IWB) is often used (Anderson et al., 2014). The study of IWB has been concerned with how individuals’ innovative actions can be measured, and which determinants impact it and in what way. 6 Acta Wasaensia The field of innovation research is vast and today comprises discussions such as open innovation, circular innovation, business model innovation, digital transformation, innovation policies, and sustainable innovation, to mention but a few of the newer streams. The need to define the scope of this thesis was evident. I chose innovative work behaviour for two reasons. First, I was interested in the relationship between certain determinants and how individuals innovate, and this type of research is often conducted under the umbrella of innovative work behaviour. Second, the measuring instruments for how individuals innovate have been developed utilising the term innovative work behaviour (e.g., de Jong & den Hartog, 2010; Janssen, 2000; Scott & Bruce, 1994), and as a former operations manager, I believe that what gets measured, gets done. Therefore, to study what aspects of individuals’ innovation processes have been measured and how seemed a task worth undertaking. To look at where these discussions take place, it can be noted that research on innovative work behaviour is published in journals of many different descriptions and also in industry-specific journals such as Tourism Management and the American Journal of Nursing. Five main interested parties can be distinguished: human resource management, management, psychology, organisational behaviour, and innovation management. Again, while many articles are general enough to have been published in any journal, some nuances are apparent. HR journals, such as Human Resource Management, have published more extensively on the role of HR practices, such as performance-based rewards, in enhancing innovative work behaviour (Sanders, Jorgensen, Shipton, Van Rossenberg, Cunha, Li, Rodrigues, Wong, & Dysvik, 2018). Management journals, such as the Academy of Management Journal, have naturally focused more on the role of the manager in the process, including whether managerial expectations of performance have an impact (Yuan & Woodman, 2010). Journals with an emphasis on psychology, such as Current Psychology, have been interested, for example, in the role of personality in innovating (Li, Liu, Liu, & Wang, 2017) whereas organisational behaviourists in journals such as Journal of Organizational Behavior have examined diverse perspectives on organisational life, such as whether innovative behaviours are typically ascribed more often to men than to women and whether the bias affects the performance evaluations of women who innovate (Luksyte, Unsworth, & Avery, 2018). Finally, journals on innovation management, such as Creativity and Innovation Management, have explored questions related to measuring innovative work behaviour (de Jong & den Hartog, 2010), among other interesting topics. Acta Wasaensia 7 2.3 Defining innovative work behaviour Creativity has typically been seen to consist of the generation of entirely novel ideas (Amabile, 1988), whereas innovation has been defined as encompassing the introduction of ideas and their application (West & Farr, 1990). The current understanding is that creativity is the domain particularly concerned with generating novel ideas that need not respond to a need or a problem, and that need not be implemented; the result of creativity is an idea. Innovation, however, always starts with an identified problem to which a solution may be an entirely new (i.e., creative) idea, or an idea applied from another context. It is also meant to be implemented. (Hughes et al., 2018). Innovation, then, has a strong practical element to it. As already stated, innovation can be studied at the individual, work team, organisational, or at multiple levels (Anderson et al., 2014). When the innovation process is studied at the individual level, it is often called innovative (work) behaviour (IWB) (Anderson et al., 2014). It refers to the different individual behaviours that are exhibited during the innovation process (Scott & Bruce, 1994). Over the years, academics have been interested in the question of whether these innovative behaviours consist of dimensions other than the idea introduction and application envisioned by West and Farr (1990), exactly how many dimensions there are, and which are the dimensions. Dividing innovation into two dimensions has been popular (e.g., Axtell, Holman, Unsworth, Wall, Waterson, & Harrington, 2000; Krause, 2004). Scott and Bruce (1994) introduced a third dimension, idea promotion, that includes behaviours related to the fact that often a new idea meets with resistance and needs to be championed to get acceptance and the resources needed to go through with the idea. De Jong and den Hartog (2010) noticed that the first dimension, idea introduction, is quite broad, and divided it into idea exploration and idea generation. Since then, even five or six dimensions have been proposed (e.g., Kleysen & Street, 2001; Lukes & Stephan, 2017). Recently, Hughes et al. (2018) analysed 159 definitions of innovation used by scholars in the past ten years and suggested that innovation is the identification of a problem, the introduction and the modification of ideas as a solution to the problem, and the promotion and implementation of those solutions. Researchers have often tried to distinguish between these dimensions of behaviour to develop ways to measure innovative work behaviour. Consequently, several different measuring instruments have been constructed, either purposefully or as a bi-product of a study (for examples of measuring instruments, see, e.g., de Jong & den Hartog, 2010; Janssen, 2000; Messmann & Mulder, 2020; Scott & Bruce, 1994). These instruments have been used to study which determinants influence 8 Acta Wasaensia the individual’s innovation process. The most popular line of research has been on the role of the manager and what can they do to ensure their employees are more innovative (for a review, see Rosing, Frese, & Bausch, 2011). Much of this research has been conducted on different leadership styles, for example, transformational leadership (e.g., Bednall, Rafferty, Shipton, Sanders, & Jackson, 2018). The study of innovative work behaviour is really the study of these three questions: What is innovative work behaviour? How can it be measured? Which factors impact innovative behaviour at work? The questions are nested, and the answer to the previous question has to be known before attempting to answer the next one. One cannot study which factors impact the innovative behaviour of employees unless one knows how to measure IWB. And one cannot measure IWB unless one knows what innovative behaviour looks like in the workplace. Figure 2 presents this visually. Figure 2. Nested nature of questions about IWB The study of innovation has generated much criticism, most of which is also applicable to the study of innovative work behaviour. The sharpest critique was presented by Anderson, De Dreu, and Nijstad as early as 2004. That research argued that the study of innovation had already become routinised, and as previous studies had identified the determinants of innovation, much of what scholars had done merely replicated and slightly extended those studies (Anderson, De Dreu, & Nijstad, 2004). Together with Potočnik and Zhou, Acta Wasaensia 9 Anderson again criticised the study of innovation for its lack of theoretical grounding and disparate approaches (Anderson et al., 2014). Hughes et al. also conducted a review into the study of leadership and innovation and called it “fragmented and primarily populated by small, ‘exploratory’ studies” (2018, p. 549). The question of how innovation is measured has been severely criticised over the past years. Under question has been, firstly, whether the nature of the measures is appropriate for the topic and why psychometric questionnaires still dominate over more experimental designs (Hughes et al., 2018). In addition, many studies utilising self-ratings has been criticised (Ng & Feldman, 2012). In their analysis of six popular measuring instruments for creativity and innovation, Hughes et al. (2018) went even deeper: they noticed that all instruments, regardless of whether they purported to measure creativity or innovation, actually measured both. In addition, the instruments also included items that measured neither concept. In other words, no instrument exists that measures exactly what it said it would measure. Furthermore, the instruments mixed items related to the innovator (e.g. traits), the innovation process, and the outcome. They conclude that we are clearly in need of “new scales that offer clear facet-level measurement and scales that distinguish between person, process, and product” since “none [of the current instruments] is particularly appropriate for future research” (Hughes et al., 2018, p. 563). Another aspect of measuring that raises concern is the dimensionality of the concept of innovative work behaviour. While most researchers theoretically distinguish multiple dimensions and agree that each involves distinct behaviours, innovative work behaviour is usually measured one-dimensionally (e.g., Newman, Tse, Schwarz, & Nielsen, 2018; Odoardi, Montani, Boudrias, & Battistelli, 2015). There have been attempts to develop a multidimensional measuring instrument (e.g., de Jong & den Hartog, 2010; Lukes & Stephan, 2017) but those have either shown insufficient construct validity, or have not been thoroughly tested (e.g., they report only Cronbach’s alphas). The reason why IWB should be measured multidimensionally is that one-dimensional measurement has produced inconsistent findings (Anderson et al., 2014). When IWB has been measured two- dimensionally, different determinants have had different effects on the dimensions (Axtell et al., 2000; Krause, 2004). It has even been suggested that some determinants might be beneficial for one dimension but detrimental to another (Perry-Smith & Mannucci, 2017). Such findings are impossible to discover using a one-dimensional measuring instrument. 10 Acta Wasaensia With regard to studying the influence of managers on innovative work behaviour among employees, research has been criticised for employing grandiose leadership styles (such as transformational or servant leadership) that do a poor job in explaining organisational reality (Alvesson & Einola, 2019; Alvesson & Kärreman, 2016). Due to the multidimensional nature of innovative work behaviour, it has been suggested that one management style may not be effective in managing the whole range of behaviours involved (Rosing et al., 2011). There have also been concerns that the effect of managerial style on innovative work behaviour has been addressed in too straightforward a manner and that, in fact, it is likely to be mediated through a mechanism (Hughes et al., 2018). Finally, as already noted, innovation can be studied at different levels. The most popular of these has been the organisational level (Crossan & Apaydin, 2010). However, in order to unpack the collective concept of organisational innovation, we have to understand how individual-level actions lead to organisation-level innovation (Felin, Foss, & Ployhart, 2015; Felin & Foss, 2005). There have been several calls for more research on how individuals generate and apply new ideas (e.g., Anderson et al., 2014; Crossan & Apaydin, 2010; Hughes et al., 2018). 2.4 Problematising the study of innovative work behaviour In this thesis, problematisation refers to exploring the problems and weaknesses of a theory relative to the phenomenon it tries to explain, allowing that which does not work in current theory to spark an interest (Alvesson & Kärreman, 2007). Although problematisation has been developed into a methodology (Alvesson & Sandberg, 2011), it is recognised that it is a creative act and as such, researchers are encouraged to find the steps that work for their problematisation process rather than follow a given script (Alvesson & Sandberg, 2011; Deacon, 2000). Problematising the study of innovative work behaviour led me to understand that there are deeper concerns with the study than those discovered before and that I detailed in the previous section. Specifically, there are three bigger issues: that the current understanding (and measurement) of innovative work behaviour hails from the 1980s and has not been updated since; that the study of behaviours is not the same as the study of the innovation process; and that there is confusion about at which level innovative work behaviour is and should be studied. Below, I will go into each issue in more detail. Acta Wasaensia 11 2.4.1 Current understanding of innovative work behaviour hails from the 1980s A study of innovative work behaviour needs to address these three questions: What is innovative work behaviour? How can it be measured? Which factors impact innovative behaviour at work? Rummaging around the literature on innovative work behaviour has led me to question whether we really know what the answer to the first question is, at least in a modern context. There is the study of which dimensions IWB consists and how it can be measured, and which determinants influence it – that is, answers to the last two questions. But searching for answers to these questions is not the same as knowing what innovative work behaviour is. Indeed, answering them is pointless if we do not know the answer to the first question, which is the foundation for the other two. The discussion around which dimensions make up IWB and how it can be measured comes close – as obviously such discussion needs to address which behaviours turn into the items in the questionnaires that ultimately measure IWB – if the discussion were based on recent empirical studies on how employees innovate at work. But I am concerned that it is not. Currently, the discussion seems to be about whether this dimension or that dimension is part of IWB based on the literature review conducted. The last studies of how employees innovate in a corporate setting might be those by Rosabeth Kanter in the 1980s (e.g., Kanter, 1988; Kanter, 1984). The informed reader may at first be sceptical: surely this is not the case! Articles about innovative work behaviour are published monthly and they extend our understanding of what innovative work behaviour is like today. My answer to this is they increase our understanding of factors that impact the type of innovative work behaviour that the measuring instruments measure. But do they add to what we know about how employees innovate in organisations today? Two separate questions need to be explored more in-depth here. One, have none of the recent studies on IWB examined how employees innovate (instead of what affects how employees innovate)? Two, are the measuring instruments all outdated? Let us take question number one first. In my literature review consisting of 255 articles utilising the keywords “innovative work behaviour” and “employee innovative behaviour” from the year 2000 until September 2020, 245 articles (or 96%) are so-called antecedent studies. Of the remaining 10 articles, only one study had done empirical research into how and why innovations are developed (Messmann & Mulder, 2011). However, the study looked at the innovative 12 Acta Wasaensia behaviour of vocational teachers, and we know that care must be taken when comparing how employees innovate in public and private organisations (Bysted & Hansen, 2015). The goals and responsibilities for vocational teachers can be very different to those of employees in private companies. Therefore, it is likely that some dimensions of innovative behaviour can carry a lot of weight for vocational teachers but are not of equal importance for employees in private companies. Likewise, dimensions that are important for employees in private companies may be missing from the scales that originate from a study of vocational teachers. As such, questions can be raised about how suitable it is to use a measuring instrument that is based on the innovative behaviours of vocational teachers in the private sector. Therefore, the answer to the first question is: no, none of the recent studies on IWB has examined how employees innovate in a corporate setting. The answer to the second question about whether measuring instruments are outdated is also found in the literature review that I conducted. That study identified 13 different measuring instruments for IWB. Most reported that they had based their instrument on previous instruments and/or literature reviews. Some articles (e.g., de Jong & den Hartog, 2010; Scott & Bruce, 1994) that had developed a measuring instrument reported having also conducted some interviews (in addition to a literature review), but the interviews in question were done at the company where the survey was later carried out, and they were more oriented to ensuring that the questionnaire was appropriate than understanding how employees innovate. Moreover, the results of these interviews were not reported, so we do not know what the researchers found. Therefore, the answer to the second question is: yes, all measuring instruments for IWB used today to measure which antecedents affect IWB are ultimately based on studies on how employees innovated in the 1980s. What are the implications of all this? Anyone who was in working life in the 1980s will, I am sure, immediately agree that things look very different in the 2020s. In the 1980s, the major question was how to get employees to bring their ideas to work after years of merely doing what they had been told to do (Kanter, 1988; Van de Ven, 1986). Dess and Picken (2000) observed that eight out of ten new jobs were for knowledge workers and that organisations and managers must therefore shift their emphasis from managing mass markets and tangible assets effectively to managing knowledge and innovation. The study recommended that leaders learn to harness the innovation capabilities of all employees in order to compete in the ‘knowledge age’. Moreover, the skills workers need have changed tremendously. The Hudson Institute predicted that in the 1990s, there would be few new jobs created for workers unable to read and understand directions, to think and speak clearly, and to do basic maths such as adding and subtracting Acta Wasaensia 13 workers (Johnston & Packer, 1987). Today, with some predicting that 50% of all work activities are automatable, workers are needed for tasks that require more advanced cognitive skills, such as creativity and complex problem solving (Manyika et al., 2017). Given that such a major chasm has appeared in working life, is it likely that the way employees behave when they innovate at work has changed since the 1980s? Yes, of course it is; yet, we use the measuring instruments that have been developed using Kanter’s observations of the IWB in the 1980s to measure the IWB of today. Of even greater concern than measuring innovative work behaviour with an outdated instrument is the fact that IWB has been measured to provide recommendations to managers on how to manage for innovation. In the best case, this advice has been sound and reliable. In the worst case, it has prompted managers to encourage behaviour that is not only inefficient in producing innovations for the organisation, but that is potentially detrimental to it. In any case, we have to know whether the innovative behaviour that Kanter observed, and that is the basis for all measuring instruments, is the behaviour that leads to innovation also in modern organisations. Directing attention to the fact that the kinds of innovative behaviours that employees should demonstrate at work today has not been studied, and presenting some tentative results of how an individual innovates at work, is one of the three issues about how IWB has been studied. 2.4.2 Behaviour is not the same as process A question related to the previous point is that of behaviour and process. The term innovative (work) behaviour is often adopted when discussing the individual innovation process (Anderson et al., 2014). But why? Why is the term individual innovation process not adopted when discussing the individual innovation process? The highly influential study (7360 citations and counting, according to Google Scholar) of Scott and Bruce (1994) was the first to use the term innovative behaviour and perhaps the phrase just stuck without anyone giving it further thought. Maybe it was thought that behaviours are easier to measure and a good enough proxy for what takes place in the individual innovation process. In any case, there are countless studies where the concepts of innovative behaviour and innovation process have been applied synonymously. In other words, innovative work behaviour has been equated with the (organisational?) innovation process 14 Acta Wasaensia when clearly, the two are not the same. As a result, while there are studies of team innovation processes (e.g., Grass, Backmann, & Hoegl, 2020) and organisational innovation processes (e.g., Damanpour & Schneider, 2009), there is no real study of individual innovation processes – only the behaviours that are necessary to the process. That the study of individual innovation processes should be concerned with the entire process and not only a part of it (i.e., behaviours), is the second issue that the study of IWB is faced with. 2.4.3 The study of innovative work behaviour is confused about the correct level of analysis Finally, I will look at the level at which innovative work behaviour has been studied, and which level it should be studied at. I will start by first discussing the types of levels that exist. The term level can refer to three things in this context: the level of theory, of measurement, and of analysis. The level of theory means the entity that generalisations are made on (e.g., organisations or individuals). The level of measurement means from which entity data are drawn. The level of analysis means the entity to which data are assigned for analysis (Mathieu & Chen, 2011). Theories can be single- or multilevel (Kozlowski & Klein, 2000). An organisational-level (O-level) single-level theory aims to theorise about organisational structures and processes or collective phenomena such as organisational culture (Devinney, 2013). An O-level single-level theory considers the behaviour of individuals to be regular (Kozlowski & Klein, 2000); that is, that the same stimulus always produces the same response. Individual-level (I-level) single-level theories focus on how the individual acts, with little regard to how the context might affect those actions (Devinney, 2013). Aggregation-level (A-level) theories incorporate multiple levels of analysis to explain how one level impacts another (Devinney, 2013). It is noteworthy that the levels are seen as impacting each other (Kozlowski & Klein, 2000). A theory that incorporates multiple levels, but considers only the higher level as able to influence the lower level, is a single-level theory. If we now consider the research conducted on innovative work behaviour, I propose that much of it is single-level and specifically O-level. Although the level of measurement is clearly individual – given that data are collected about employees’ innovative behaviours – the individual is expected to react in a predictable and regular manner, typical of methodological collectivism (Agassi, Acta Wasaensia 15 1960). Therefore, previous research has represented more single-level than multilevel theorising. This may have something to do with the fact that creativity was originally thought to occur at the individual and team level, and innovation at the organisational level (Amabile, 1988). Additionally, when the foundations for the study of innovative work behaviour were laid (e.g. de Jong & den Hartog, 2010; Janssen, 2000; Scott & Bruce, 1994), organisation theory was strongly focused on macro explanations (Felin et al., 2015; Felin & Hesterly, 2007). That one could and should explain macro concepts at the micro level was not a popular argument and led, perhaps, to some scholars attempting to “raise” their level of analysis to the macro level. The need for a multilevel approach is clear (Anderson et al., 2014; Mathieu & Chen, 2011). Such an approach would include multiple levels of analysis and consider those levels as capable of influencing each other; that is, that the lower level (that of the individual) can influence the organisational level and not only the other way around (Kozlowski & Klein, 2000). Current studies of innovative work behaviour examine innovative work behaviour as the dependent variable, as if the buck stopped there when, of course, it does not. Innovative work behaviour is not the end stop, it is a means to an end. Often, that end is innovative outcomes but it could also be other things such as improved team commitment or organisational culture. This ultimate end result is often implied in current research and references are made to previous studies that have verified that innovative work behaviour leads to innovations, but although some exceptions exist (e.g., Gambi, Lizarelli, Junior, & Boer, 2021), insufficient attention has been paid to the study of what the outcomes of innovative efforts are (Janssen et al., 2004). The third issue that I found with the study of IWB is unclarity around which level of theorising has so far been employed. My recommendation is that as scholars of individual innovation, we honestly conduct research at the individual level to help explain organisational level phenomena. The microfoundations movement provides solid reasons for doing so, given that its aim is “to unpack collective concepts to understand how individual-level factors impact organizations, how the interaction of individuals leads to emergent, collective, and organization-level outcomes and performance, and how relations between macro variables are mediated by micro actions and interactions” (Felin et al., 2015, p. 576). In the past decade or so, the microfoundations movement has had substantial success (Felin et al., 2015) and has made explaining collective concepts with the help of individual-level factors acceptable. In fact, some have gone as far as stating that “truly explaining …the organization (e.g. existence, decline, capability, or performance), or any collective for that matter, requires starting with the individual as the central actor” (Felin & Foss, 2005, p. 441, emphasis added). 16 Acta Wasaensia To be clear, I am not claiming that innovation is mostly a solitary activity. On the contrary, I acknowledge that important research has been conducted on team innovation (Alexander & Van Knippenberg, 2014; Mitchell & Boyle, 2015; Tang, Chen, van Knippenberg, & Yu, 2020) and my own research also pointed to innovation being a team effort. What I am saying, though, is that it also important to study the actions of individuals in a team without necessarily aggregating to team level in order to preserve some of the heterogenous voices of the individuals (Barney & Felin, 2013). 2.5 Objective and research questions Initially, my aim was to improve the understanding of how organisations can remain innovative, having seen first-hand the benefits to both the organisation and to its employees. Having been in a managerial position myself and having a strong desire to contribute to the training and development of better managers, I was keen to understand the role of managers in fostering individual innovation. As I explored the topic I entered into a hermeneutic circle where the more I understood about the topic, the more questions I had, and the deeper I had to dig. Table 1 describes how the results of the previous paper shaped the research questions for the next paper. In addition to the individual research questions that each of the papers answered, as presented in Table 1, the hermeneutic circle also produced an overarching research problem that I explored in the entire dissertation: How should the study of innovative work behaviour be developed to respond to the needs of modern working life? Although seemingly simple, the problem is multifaceted. First, to know how the study should be developed one must know what the current state is, and that there is room, and a need, for development. It is also necessary to understand whether the way that the study is currently conducted meets the needs of modern working life; that is, to what extent does the study portray the phenomenon as it exists in the world today. Finally, there is the question of which the most suitable ways are to develop the IWB study. Due to these many aspects, the research problem invites theoretical, methodological, and practical perspectives. I will return to these in Chapter 6. Acta Wasaensia 17 Table 1. Description of the hermeneutic research process Research questions Main results Questions that arose Paper 1 • Does managerial coaching (MC) have a positive relationship with IWB? • Does MC have the same effect on all four dimensions of IWB? • MC is an appropriate tool for fostering IWB • When IWB is measured four- dimensionally, the effect of MC was observed to grow throughout the process • Does MC affect IWB through a mediating mechanism? • What issues are there in measuring IWB multidimensionally? Paper 2 • Does MC affect IWB through work engagement (WE)? • Is the effect the same in all four dimensions of IWB? • The effect of MC on IWB is stronger through WE than directly • The effect grows throughout the process • Is the current picture of IWB up- to-date? • Are we really measuring the IWB of today? Paper 3 • What is studied about innovative work behaviour today? • How have the current measuring instruments been developed? • 97% of studies since 2000 are replication- extension studies • The instruments used for measuring IWB are based on the work by Kanter in the 1980s • How has the individual innovation process changed from how previous literature has conceived it? Paper 4 • How has the individual innovation process changed from how previous literature has conceived it? • The innovation process consists of different activities than described in previous literature in our case study • The role of the individual in the innovation process is more active and engaged than previously thought of • What does the individual innovation process look like in other contexts? • Do managers and team members experience the innovation process the same way? • Are behaviours the right construct to measure? • To what extent is it possible to theorise on the individual innovation process so that it applies to all or most contexts? 18 Acta Wasaensia 2.6 Thesis structure The thesis comprises six chapters that together form a background for the four individual papers that make up the second part of the dissertation. The introductory chapter has provided brief background information on the topic, and outlined the need for this study and its objectives. The next chapter delves deeper into the concept of innovative work behaviour: what it is, and what is currently known about it. The fourth chapter is dedicated to the methodological choices and assumptions made in this dissertation. The fifth chapter presents a summary of the four appended papers and their contributions. The results shed light on what tangible levers managers have available to encourage their employees to be more innovative, and on the nature of the phenomenon itself: what innovative work behaviour looks like today, and how it has been measured. The sixth and final chapter discusses the contributions of this entire thesis theoretically, methodologically, and practically, before reflecting on the limitations and directions for further research. Acta Wasaensia 19 3 THEORETICAL BACKGROUND This chapter presents the theoretical background of the thesis. First, I adopt a broader perspective on the study of innovation: what has been studied and at what level, and how is innovation different from creativity. As the backbone, I have utilised three highly regarded systematic literature reviews: those by Anderson et al. (2014), Crossan & Apaydin (2010), and Hughes et al. (2018), and added to them. The scope is then narrowed to the level of the individual, and I look at what kind of study is currently being conducted on how individuals innovate. Here, I summarise the results of the systematic literature review that I conducted for Paper 3. I also discuss dimensions of innovative work behaviour, measuring instruments, and managing individual innovation. A look at the current state of science of research on individual innovation completes the chapter. 3.1 The study of innovation The purpose of this section is to position this thesis in the wider innovation literature. To help do this, I compiled Figure 3 which combines two categorisations of innovation study: at what level the study is conducted, and what stage of innovation the study is concerned with. The study of innovation can take place at the individual, work team, organisational, or multiple levels (Anderson et al., 2014), as presented in the horizontal bars in Figure 3. The vertical bars show an alternative categorisation that distinguishes three streams of innovation research: one that focuses on the determinants of innovation, one that looks at innovation as a process, and one that is interested in innovation outcomes (Crossan & Apaydin, 2010). The streams are related: determinants affect the innovation process, and the innovation process precedes innovation outcomes (Crossan & Apaydin, 2010). The two dots in Figure 3 represent the focus areas of this thesis and the line indicates that I was, at least to start with, interested in exploring how organisational-level determinants affect the individual innovation process. As far as I am aware, I am the first to combine the two categorisations. The benefit of doing so and presenting the result visually is that it makes it easier to grasp what is being talked of. It gives a framework for structuring literature on innovation and allows for observations about which areas have received the most attention, and which are so far under-researched. Innovation studies can be mapped into this framework and with the help of dots and lines, it can be made clear whether a study explores one of the streams (and at which level) or whether the focus is on the effects of one stream on another (and at which level). To make such an intent clear 20 Acta Wasaensia in the beginning of a study would be beneficial both to the readers and to the scholars themselves to avoid ambiguity. The framework immediately shows that at least theoretically, it is possible to study determinants of innovation, the innovation process and its outcomes at all the three different levels but also that innovation scholars may want to challenge the framework and research questions outside of it. In Section 6.2, I will discuss some opportunities for further study that this presents. In what follows, I will go through each of the streams and what has been studied so far at the different levels. Figure 3. Initial research interest 3.1.1 Innovation outcomes Starting with innovation outcomes, the majority of research has taken place at the organisational level (Crossan & Apaydin, 2010). The outcome is generally either radical or incremental innovation (Gopalakrishnan & Damanpour, 1997) related to products or services, processes, or business models, and new either to the firm, its market, or the entire industry (Crossan & Apaydin, 2010). Recent interest areas in this body of research have included, for example, how business model innovation takes place in strategic alliances (Spieth, Laudien, & Meissner, 2021). Acta Wasaensia 21 At the other levels (team and individual), relatively little research exists on what could be considered specifically team or individual outcomes of innovation; the outcome of team or individual efforts is usually expected to be organisational-level innovation. Additionally, most studies at the team and individual level have focused on determinants of innovation and have therefore examined innovation as a dependent variable, and studies utilising the innovation process as an independent variable are scarce (Janssen et al., 2004). Some such studies do exist and have shown that the outcome may be positive, such as enhanced employee engagement and well-being (Huhtala & Parzefall, 2007); however, outcomes can also be negative – for example, risk of conflict and resistance to change from other people when pushing through innovative ideas (Janssen, 2003). A recent study finds a positive outcome: engaging in innovative behaviours shields employees from burnout after company downsizing (Hammond et al., 2019). 3.1.2 Innovation processes Innovation processes refer to the activities and interactions required for an innovation to be generated and implemented. As mentioned before, when studied at the individual level, the term innovative (work) behaviour is often used (Anderson et al., 2014). This is one of the two focus areas in this thesis and will be discussed in more detail in Section 3.2. At the team level of analysis, Anderson et al. (2014) in their comprehensive review were surprised by how few studies examine within-team processes. Since then, more and more research has taken the team-level view on innovation. For example, a model of team innovation process was recently developed that focuses on empowerment as the key human factor within agile teams (Grass et al., 2020). Alexander and van Knippenberg (2014) examined the success factors behind teams pursuing radical innovation and found that switching team goal orientations may increase success in radical innovation. At the organisational level, the discussion has tended to start with the distinction between creativity and innovation. In some of the most influential work in this arena, Amabile (1988) conceptualised creativity as novel and useful ideas produced by an individual or a small group, whereas innovation was presented as an organisational concept ensuring that the creative ideas are successfully implemented; the two concepts are related but clearly differ from each other. According to West and Farr (1990), the innovation process consists of both idea generation (i.e., creativity) and the application of ideas, where creativity is the initial step in the innovation process. West and Farr provoked a lively debate on the exact nature of creativity and innovation that features arguments for creativity being the first step in the innovation process (Černe, Hernaus, Dysvik, & Škerlavaj, 22 Acta Wasaensia 2017) and also stand-alone concept (Anderson et al., 2014). In practice, the two concepts are often confused and even top-tier journals have published articles referring to innovation but citing sources from the creativity literature (Hughes et al., 2018). While the innovation process has always been seen as consisting of different phases, different strands of literature have conceptualised the phases differently. For example, Damanpour and Schneider (2006) distinguish between initiation, the adoption decision, and the implementation phases. Amabile and Pratt (2016) suggest five phases: agenda setting, stage setting, producing ideas, testing and implementing ideas, and outcome assessment. One stream of research starts with Kanter who distinguished four innovation ‘tasks’ that were described as “correspond[ing] roughly (but nowhere near exactly) to the logic of the innovation process as it unfolds over time” (Kanter, 1988, p. 96). These tasks are idea generation, coalition building, idea realisation, and the diffusion of knowledge in commercialisation of the product (Kanter, 1988). It is this stream that is followed in this thesis. Similar conceptualisations have since been formulated by scholars including Scott and Bruce (1994), Janssen (2000), and de Jong and den Hartog (2010), and Section 3.2 will elaborate further on this. Hughes et al. (2018) have argued the importance of clearly defining, and making a distinction between, the concepts of creativity and innovation. In their review of 195 articles concerned with the effects of leadership on creativity and innovation, they coded all definitions of creativity and innovation found in their article cache. They then formulated their own definition based on the codings. They disregarded any codings mentioning organisational benefits or useful ideas, arguing that a phenomenon exists outside of its effects; that is, an idea can be innovative even if it is not yet known whether it has any benefits. What remained was that the overwhelming majority (96%) of creativity definitions agree that creativity is the generation of entirely new or original ideas. They distinguished five conceptual markers in the definitions of innovation: problem recognition, introducing and modifying (relatively) new ideas, promoting the ideas, and finally implementing them. (Hughes et al., 2018). Based on this, they suggest that the two key differences between creativity and innovation are, 1) only creativity refers to the generation of brand new ideas; innovation refers to ideas that are new to the context but not necessarily to the world; and 2) innovation is born out of a need or as a solution to a problem and is always implemented, whereas creative ideas can occur with no specific need in mind, nor do they need to be implemented. Acta Wasaensia 23 3.1.3 Determinants of innovation Finally, when talking about determinants (the terms antecedents and predictors are also used) at the individual level, these are factors that an individual possesses or that are at least to some degree within an individual’s control. Such factors include individual differences (such as personality and traits), motivation, knowledge and abilities, and psychological states (Anderson et al., 2014; Hammond, Neff, Farr, Schwall, & Zhao, 2011). At the team level, a meta-analysis of team-level predictors found that support for innovation, vision, task orientation, and external communication all had strong relationships with innovation, whereas team structure and composition did not (Hülsheger, Anderson, & Salgado, 2009). Researchers at Google studied 699 people working on group tasks and discovered that what distinguished the well-performing teams was that everyone spoke approximately the same amount, and the team members were good at picking up social cues from one another (Bergmann & Schaeppi, 2016; Duhigg, 2016). Organisational-level determinants influence not only the organisational innovation process but also team and individual processes. These determinants can be divided into five groups (drawing inspiration from Anderson et al., 2014; Crossan & Apaydin, 2010): the characteristics of the innovation itself, factors related to the external environment, organisation-wide factors, factors concerned with top management, and factors within an individual manager’s control. Innovation characteristics such as its cost, how complex it is, and how advantageous it will be for the organisation influence innovation adoption (Damanpour & Schneider, 2009). Different aspects of the external environment include factors such as urbanization and the unemployment rate (Damanpour & Schneider, 2006). Organisation-wide factors may include the characteristics of the organisation such as its size and complexity (Damanpour & Schneider, 2006) or its practices such as organisational climate (Shanker, Bhanugopan, van der Heijden, & Farrell, 2017), HR practices (Bos-Nehles, Renkema, & Janssen, 2017) such as performance-based rewards (Sanders et al., 2018), strategy (Adams et al., 2006), and knowledge management (Battistelli, Odoardi, Vandenberghe, Di Napoli, & Piccione, 2019). With regard to factors related to top management, previous research has looked at CEO demographics such as tenure (Wu, Levitas, & Priem, 2005) and diversity in occupational background at the board level (Goodstein, Gautam, & Boeker, 1994). The different leadership styles of CEOs have also been studied; for example, a recent study found that entrepreneurial leadership encourages the innovative behaviours of employees with creative self- efficacy and passion for inventing as mediators (Bagheri, Newman, & Eva, 2020). 24 Acta Wasaensia The fifth and last category of determinants are factors within an individual manager’s control, which is one of the focus areas of the thesis. Accordingly, that category is examined in some detail in Section 3.4. 3.2 The individual innovation process Studying organisational innovation has been by far the most popular course of action for innovation researchers: in their systematic review of 525 articles, Crossan and Apaydin (2010) found that 52% of the articles explored innovation at the organisational level and only 5% did so at the individual level. Nevertheless, there appear to remain significant opportunities to improve the understanding of macro level outcomes through investigating their microfoundations since individuals make up organisations and organisations do not exist without individuals (Felin et al., 2015; Felin & Foss, 2005). Furthermore, while the organisational innovation process has been presented as consisting of phases such as idea generation and idea promotion, it is individuals who do the generating and promoting (Scott & Bruce, 1994; Van de Ven, 1986). Therefore, in this thesis, the actions of individual actors in organisations receive primary attention. When the innovation process is studied at the individual level, it is often called innovative (work) behaviour (IWB) (Anderson et al., 2014). The fact that research on the individual innovation process is, in fact, research on individual behaviours in the (organisational?) innovation process, is significant, and not without its complications. It means that the individual innovation processes are actually not being studied at all; behaviours within the process are. Some researchers have stated explicitly that process and behaviour are not the same. Kanter (1988) talked about four innovation tasks, which she roughly equated to the innovation process. Scott and Bruce (1994, p. 582) viewed innovation “as a multistage process, with different activities and different individual behaviors necessary at each stage”. In later research, the words process and behaviour are sometimes used interchangeably or behaviours are talked of as having stages (e.g., Stoffers, Van der Heijden, & Jacobs, 2020), indicating that the concepts have become muddled. In this thesis, I use the term innovative work behaviour in Papers 1–3, and the term individual innovation process in Paper 4, which looks at the whole process instead of only behaviours. Janssen (2000) noted that since generating, promoting, and implementing ideas for the improvement of the organisation are not in a regular employee’s job description, innovative behaviour can be classified as extra-role behaviour. Indeed, in an era where a mental shift had to be made from obeying to bringing Acta Wasaensia 25 idea power to work (Kanter, 1984), this was probably the case. Now, innovative behaviour is often seen as a specific form of performance (Montani, Vandenberghe, Khedhaouria, & Courcy, 2020), and part of an employee’s regular duties. Innovative behaviour is thus separate from employee-driven innovation; the latter specifically relating to innovation beyond the normal job description of an employee (Bäckström & Bengtsson, 2019; Kesting & Ulhøi, 2010). One of the first mentions of the concept of innovative behaviour is by Scott and Bruce (1994), who developed and tested a model of innovative behaviour on the individual level. They drew on the work of Kanter (1988) to conceptualise innovative behaviour as consisting of recognising problems, generating ideas and solutions, seeking sponsorship for the idea and building a coalition to support it, and producing a prototype of the innovation. Innovative behaviour, then, has always been thought to be a multidimensional concept and the question of how many dimensions there are and what they are interests scholars to this day. Some of the suggestions are two (Krause, 2004), three (Janssen, 2000; Scott & Bruce, 1994), four (de Jong & den Hartog, 2010), five (Hughes et al., 2018; Kleysen & Street, 2001), or even six (Lukes & Stephan, 2017) dimensions. Table 2 summarises the different dimensions suggested in some of the previous literature. We can see that when two dimensions are distinguished, they are related to generating ideas and implementing them (Krause, 2004). When a third dimension is added, as Scott and Bruce (1994) and Janssen (2000) did, idea promotion becomes a dimension on its own. De Jong and den Hartog (2010) separated idea generation into two dimensions with idea exploration preceding the generation of ideas; a conceptualisation that is fairly close to the one proposed by Kanter (1988). Kleysen and Street (2001) divided the early stages even further and recognised that after the problem or opportunity has been explored and ideas for it generated, the ideas usually go through some type of refinement or development. Lukes and Stephan (2017) are the only ones who divided the idea implementation stage further. In their model, ideas are first searched for and generated, after which they are communicated. Implementation starting activities follow (making plans, essentially), after which others need to be involved in the implementation and coalitions built. Throughout the implementation process, Lukes and Stephan saw that obstacles would emerge that must be overcome (Lukes & Stephan, 2017). Hughes et al. (2018) coded all the definitions of innovation they identified in their article cache of 195 articles, spanning roughly the previous ten years. The work reviews previous literature and suggests that five dimensions (problem recognition, idea introduction, idea modification, promotion, and implementation) have often been used by innovation researchers. Some 26 Acta Wasaensia dimensions have been included in definitions more often (e.g., the implementation dimension had been mentioned in 75% of the definitions), and others less often (e.g., problem recognition was present in 4.4% of the definitions) (Hughes et al., 2018). The review provides a handy snapshot of which dimensions are generally thought to make up innovative behaviour, and is highlighted in Table 2 in italics to indicate that their dimensions are the result of a review of other researchers’ definitions. Table 2. Previously suggested dimensions of IWB Krause, 2004 Scott & Bruce, 1994; Janssen 2000 de Jong & den Hartog, 2010 Kleysen & Street, 2001 Hughes et al., 2018 Lukes & Stephan, 2017 Gene- ration & testing of ideas Idea gene- ration Idea exploration Opportunity exploration Problem recognition Idea search Idea generation Generativity Idea introduction Idea generation Formative investigation Idea modification Idea promotion Idea championing Championing Idea promotion Idea communication Imple- mentation of ideas Idea realisation Idea imple- mentation Application Idea implementation Implementation starting activities Involving others Overcoming obstacles Another question about how managers can enhance the innovative work behaviour of their employees that remains open is whether all dimensions of innovative behaviour require the same antecedents. Innovative behaviours vary and change depending on the phase of the innovation process. When an employee looks for opportunities, their behaviour is of a certain kind; for example, they might ask questions about a specific product to determine whether its performance could be improved. When an employee wants to sell their performance improvement idea to their manager (to obtain money and other resources), the behaviours are different and might include calculating the return on investment for the innovation. Do the same factors help the innovator in these two phases? Some research suggests that different factors, and different managerial support, are needed (e.g., de Jong & den Hartog, 2007; Fang, Chen, Wang, & Chen, 2019; Perry- Smith & Mannucci, 2017). Acta Wasaensia 27 3.3 Measuring innovative work behaviour For as long as the concept of innovative work behaviour has existed, it has been measured. It has been measured mainly to acquire information about which antecedents – the factors within the manager’s or the organisation’s control – have an impact on it and can thus be used to enhance the innovative behaviours of employees. I will delve deeper into this topic in the next section but for now, I will focus on what types of measuring instruments exist and identifying some of the known issues with them. To the best of my knowledge, the first measuring instrument that attempted to measure innovative behaviour was developed by Hurt, Joseph and Cook (1977). Work on developing a measuring instrument for innovative work behaviour really got underway with the instrument developed by Scott and Bruce (1994). Since then, several measuring instruments have been developed and the work continues to this day; two proposals for new instruments emerged only recently (Lambriex- Schmitz, Van Der Klink, Beausaert, Bijker, & Segers, 2020; Messmann & Mulder, 2020). Instrument development work has closely mirrored the work on dimensions of innovative behaviour (described in Section 3.2). This is because work on how innovative behaviour can be measured has necessarily always included figuring out what it is that should be measured. In the systematic literature review that I conducted, I found that articles examining different aspects of innovative work behaviour since the year 2000 reported using more than 30 different measuring instruments for IWB. A closer inspection of the instruments revealed colourful practices in what is considered a measuring instrument; it was quite customary, for example, to report using a measuring instrument that was developed in a certain article but when reading said article the author(s), in turn, reported using a measuring instrument developed by someone else. Having followed these trails to the end led to 13 unique measuring instruments that can, with some confidence, be considered a relatively accurate number of measuring instruments for IWB that researchers have used in the past 20 years. These instruments can be divided into three categories based on whether they measure innovative work behaviour one- or multidimensionally. In the first category, the developers theoretically distinguish more than one dimension but they do not even attempt to measure innovative work behaviour multidimensionally. Such cases are for example Scott and Bruce (1994) and Radaelli, Lettieri, Mura and Spiller (2014) who see the innovation process as consisting of idea generation, promotion, and implementation but whose 28 Acta Wasaensia instruments measured the concept one-dimensionally in the article where the instrument was introduced. In the second category, the instrument was designed to be multidimensional but when tested, it did not show sufficient validity (i.e., the dimensions were shown to all belong under the umbrella concept of innovative work behaviour but they were not distinct enough from one another; the boundaries between idea exploration and idea generation, for example, were blurred) and therefore the developers reverted to measuring innovative behaviour one-dimensionally. Examples of these types of measuring instruments are the ones by Kleysen and Street (2001) and de Jong and den Hartog (2010). The third category is for measuring instruments that sought to measure innovative work behaviour, did so, and reported that they succeeded. A closer examination of these instruments reveals that none is particularly good for wider use. Some of the instruments reported Cronbach’s alphas and item loadings for the factors. However, they did not report conducting tests for scale validity and consequently we do not know whether the dimensions were distinct enough from one another (Axtell et al., 2000; Krause, 2004). Some instruments conducted more thorough testing but consisted of 35 items (Messmann & Mulder, 2012) and 44 items (Lambriex-Schmitz et al., 2020) which does not make them very user-friendly; additionally, these two instruments were developed specifically for measuring the innovative work behaviour of teachers. Issues with measuring instruments Measuring innovative work behaviour multidimensionally has evidently proved difficult. Successfully measuring IWB multidimensionally is, however, important for two reasons. First, when IWB is measured one-dimensionally the results tend to be inconsistent. There is disagreement on for example the role of such managerial practices as rewards or leadership styles (Anderson et al, 2014) and whether it is better to work on innovations alone or as a team (Perry-Smith & Mannucci, 2017). A likely explanation is that some parts of the innovation process are best worked on alone and others with other people but a one-dimensional measuring instrument cannot capture the difference. Second, when innovative work behaviour has been measured two-dimensionally (Axtell et al., 2000; Krause, 2004; Veenendaal & Bondarouk, 2015), the antecedents studied have had a different effect on the dimensions. For example, when de Jong and den Hartog (2007) examined the effects of 13 different leadership behaviours on two dimensions of innovative work behaviour, they found that some behaviours Acta Wasaensia 29 impacted idea generation more, whereas others had a bigger effect on idea implementation. It has even been suggested that an antecedent that is beneficial for one dimension can be detrimental to another (Perry-Smith & Mannucci, 2017). With regard to the measuring instruments themselves, it is clear that most instruments measure innovative work behaviour using a survey instead of, for example, a divergent thinking test (Hughes et al., 2018), and in many studies, self- ratings are used which often result in larger effect sizes (Ng & Feldman, 2012). Hughes et al. (2018) conducted a thorough review of six of the most commonly used measuring instruments for creativity and innovation and found significant issues. First, the scales mix items related to the personality of the innovator (e.g., traits), the innovation process, and innovative outputs. Second, scales that purport to measure innovative work behaviour often also include items that measure creativity or something else entirely that is neither innovation nor creativity. Third, most of the instruments failed to demonstrate scale accuracy through appropriate psychometric analyses. In addition, concerning the use of measuring instruments (by researchers other than those who had developed the scale), Hughes et al. (2018) reported that it was common to take a subset of items from a scale, or combine items from several scales without conducting thorough analyses to ensure the reliability and validity of the newly formulated scale. In Paper 3, I add my own observations relating to the issues with current measuring instruments and find that we may have cause to question whether any of the scales measure the innovative work behaviour of today. When carefully examining the development methods of 13 measuring instruments, I discovered that with the exception of one measuring instrument (Messmann & Mulder, 2012), all were developed based on previous literature alone; no one has examined what the individual innovation process looks like since the work done by Kanter in the 1980s. The case study, the results of which are reported in Paper 4, provides proof that an individual’s innovation process looks different from how current measuring instruments depict it. These results are explained in more detail in Section 5.4. 3.4 Managing innovative work behaviour In Section 3.1.3, several different types of determinants at different levels of analysis were identified. Many different types of factors have been shown to have an effect on employees’ innovative behaviour and choosing any one of them would have been justified in this thesis. However, the focus here is on the ways and means available to a line manager or equivalent (a separate group from top management) 30 Acta Wasaensia to influence the innovative behaviour of employees, for two reasons. First, leaders have been characterised as key drivers of innovative behaviours (Krause, 2004; Schuh, Zhang, Morgeson, Tian, & van Dick, 2018) and their role to the innovation process has been the topic of a growing number of studies (Hughes et al., 2018). Second, leadership effect has mostly been studied using traditional management models, and employing managerial coaching as the leadership influence in Papers 1 and 2 allowed me to choose a more modern approach whose effect on IWB is not yet well-studied. The way leaders impact on individual-level innovation can be divided into two categories: leadership styles and supervisory practices. We will take a look at both here since managerial coaching can include elements of both. 3.4.1 Leadership styles Leadership style is a key predictor of innovation, and the effects on individual innovation of several styles have been studied. One of the most studied styles is the transformational leadership style (Hughes et al., 2018). In general, transformational leadership has been found to relate significantly to innovative work behaviour (Choi, Kim, Ullah, & Kang, 2016; Reuvers, Van Engen, Vinkenburg, & Wilson-Evered, 2008). It has been suggested that the relationship between transformational leadership and innovative work behaviour is non-linear; that the positive effect of transformational leadership is stronger at high and low levels and weaker in the middle (Bednall et al., 2018). The effects of both transformational and transactional leadership have also been studied; when the followers exhibited high levels of psychological empowerment, transformational leadership positively influenced innovative work behaviour while transactional leadership had a negative effect (Pieterse, van Knippenberg, Schippers, & Stam, 2009). Others have found that one component of transactional leadership, verbal rewards, has a positive influence on innovative behaviour (Hansen & Pihl- Thingvad, 2019). Leader-member exchange theory (LMX) has also been a popular leadership style to study as an influencer of innovative work behaviour and has been found to lead to psychological empowerment, which in turn prompts more innovative behaviour (Schermuly, Meyer, & Dämmer, 2013). Another study found that LMX has a positive and significant effect on innovative behaviour when there are fewer within-group strong ties, and also to fully mediate the effect that out-group weak ties have on innovative behaviour (Wang, Fang, Qureshi, & Janssen, 2015). Employees with high-quality LMX relationships also consistently get favourable Acta Wasaensia 31 performance ratings when they exhibit innovative work behaviour (Schuh et al., 2018). Other leadership styles studied in connection with innovative work behaviour include the empowering, authentic, and servant leadership styles. All have, in general, been found to be positively related to IWB (Hughes et al., 2018). As is true of the other leadership styles, research has lately focused on how the leadership styles moderate or mediate the relationship between some other determinant and innovative work behaviour; for example, that servant leadership mediated between ethical climate perception and IWB (Topcu, Gursoy, & Gurson, 2015). Another, more popular trend is to examine which factors might mediate or moderate the relationship between a certain leadership style and IWB; an example is the finding that psychological empowerment moderates the relationship between authentic leadership and IWB (Grošelj, Černe, Penger, & Grah, 2020). While it has been far more common to examine the effects of positive leadership styles on IWB, a few studies have also looked at more negative styles. As expected, a destructive leadership style negatively influences millennial employees’ IWB (Hou, 2017), as does abusive supervision (Wang, Li, Zhou, Maguire, Zong, & Hu, 2019). 3.4.2 Supervisory practices In this section, we only look at practices under an individual supervisor’s control, and therefore solely organisation-wide HR practices are excluded. More research has been conducted on leadership styles than on supervisory practices. The studies focusing more on practices have often explored the role of feedback and found that feedback from supervisors moderates the positive relationships between time pressure and skill variety and innovative work behaviour (Noefer, Stegmaier, Molter, & Sonntag, 2009). Feedback is also directly and positively related to innovative work behaviour, and that relationship is mediated by work engagement and perceptions of breaches of a psychological contract (Eva, Meacham, Newman, Schwarz, & Tham, 2019). If the managers let their employees know that innovative behaviours are expected of them, employees engage in them more frequently (Yuan & Woodman, 2010). Some participative practices have also been studied. Supervisors who permit their employees autonomy and encourage skill development can encourage innovative behaviours among those employees (Bysted & Jespersen, 2014). Supervisors who help employees develop their proactive goal-setting skills are likewise likely to provide a driver of innovative behaviour at work (Montani, Odoardi, & Battistelli, 32 Acta Wasaensia 2014). Facilitating knowledge sharing is also a practice worth fostering (Kim & Park, 2017). Managers can, at least to some extent, constrain innovation. A review found that a moderate level of input constraints (such as time, money, and equipment) motivates experimentation by framing the task as a challenge. However, both too high and too low levels of constraint are detrimental to innovation (Acar, Tarakci, & van Knippenberg, 2019). 3.4.3 Issues with managerial determinants The use of these determinants has not escaped criticism. First, leadership styles such as transformational leadership and LMX have been seen as too grand and heroic to truly study organisational reality (Alvesson & Einola, 2019; Alvesson & Kärreman, 2016). Second, the nature of the innovation process, which encompasses both creative (exploratory) elements and implementational (exploitative) elements spurs concerns over whether any one leadership style is likely to be enough (Rosing et al., 2011). Third, some scholars strongly recommend abandoning current approaches to measuring charismatic-transformational leadership styles due to fundamental issues with its conceptualisation and measurement validity (van Knippenberg & Sitkin, 2013). Additionally, Anderson et al. warned in 2004 that innovation research had become routinised and that too many replication-extension studies had been published already then (Anderson et al., 2004). The systematic literature review that I conducted found that the situation since then has only exacerbated and that 96% of the 255 articles published on innovative work behaviour since the year 2000 are quantitative studies exploring the effect of one determinant or another on innovative work behaviour. The research settings have also become more complex over time, probably because the direct effects of the most obvious determinants (such as different leadership styles, as already discussed) have already been studied. Moderated mediation analyses have become popular, an example of which is one investigating the relationship between top-down knowledge hiding and IWB with self-efficacy first as a mediator and further moderated by the nationality of supervisor and employee (Arain, Bhatti, Hameed, & Fang, 2019). Another example of a fairly complex research setting is a multilevel study utilising two sources. That study found two- and three-way interactions with decision autonomy, task interdependence, and a mastery climate moderating the relationship between knowledge hiding and innovative work behaviour (Černe et al., 2017). Acta Wasaensia 33 A further issue is the continued practice of measuring innovative work behaviour one-dimensionally, despite the method tending to produce inconsistent findings. Only a few studies exist that examine the effects of different types of managerial determinants on two dimensions of innovative work behaviour. Noefer et al. (2009) discovered that supervisor