Entrepreneurial Individuals: Empirical Investigations into Entrepreneurial Activities of Hackers and Makers

Halbinger, Maria Doctoral Thesis Entrepreneurial Individuals: Empirical Investigations into Entrepreneurial Activities of Hackers and Makers PhD Series, No. 15.2014 Provided in Cooperation with: Copenhagen Business School (CBS) Suggested Citation: Halbinger, Maria (2014) : Entrepreneurial Individuals: Empirical Investigations into Entrepreneurial Activities of Hackers and Makers, PhD Series, No. 15.2014, ISBN 9788793155312, Copenhagen Business School (CBS), Frederiksberg, https://hdl.handle.net/10398/8931 This Version is available at: https://hdl.handle.net/10419/208891 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by-nc-nd/3.0/ Maria Halbinger The PhD School of Economics and Management PhD Series 15.2014 PhD Series 15.2014 Entrepreneurial Individuals copenhagen business school handelshøjskolen solbjerg plads 3 dk-2000 frederiksberg danmark www.cbs.dk ISSN 0906-6934 Print ISBN: 978-87-93155-30-5 Online ISBN: 978-87-93155-31-2 Entrepreneurial Individuals Empirical Investigations into Entrepreneurial Activities of Hackers and Makers Entrepreneurial Individuals Empirical Investigations into Entrepreneurial Activities of Hackers and Makers Maria Halbinger PhD School in Economics and Management Copenhagen Business School Maria Halbinger Entrepreneurial Individuals Empirical Investigations into Entrepreneurial Activities of Hackers and Makers 1st edition 2014 PhD Series 15.2014 © The Author ISSN 0906-6934 Print ISBN: 978-87-93155-30-5 Online ISBN: 978-87-93155-31-2 “The Doctoral School of Economics and Management is an active national and international research environment at CBS for research degree students who deal with economics and management at business, industry and country level in a theoretical and empirical manner”. All rights reserved. No parts of this book may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or by any information storage or retrieval system, without permission in writing from the publisher. i ACKNOWLEDGEMENTS To my parents Creativity And passion for what you do Make your dreams come true This PhD was for me the unique opportunity to explore a topic of personal interest and a period where I experienced wonderful support from faculty, friends and family. To these individuals, I would like to express my sincere appreciation. First and foremost, I would like to thank my main advisor Professor Toke Reichstein. Thank you, Toke, for being so very faithful in my work from the beginning, for boosting my confidence and giving me the freedom to collect and analyze an “exotic” dataset, and for being an invaluable sparring partner, friend and mentor – and of course, for divulging the Holy Grail of econometrics to me. Working and writing with you has been clearly one of the most enlightening learning experiences and most enjoyable work experiences in my PhD years. Professor Thomas Roende has been my second strong and supportive advisor throughout my PhD. Thank you, Thomas, you have been a great source of advice. You always took time to help me out and I could always count on your well-reasoned opinion at crucial decision-points, including the job market phase. I wish to express a sincere thank you to my dissertation committee, Professor Mark Lorenzen, head of committee, Professor Maryann Feldman and Professor Oliver Alexy for generously offering their time and constructive feedback. This is a great honor and the committee’s support has been absolutely invaluable. ii Professor Ulrich Kaiser, together with Mark Lorenzen built my pre-defense committee. The in-depth discussion of my thesis and their comments helped me significantly to prepare this dissertation for the final submission. Thank you very much. I also gratefully acknowledge the support I have received from the Regensburg hackerspace and Labitat hackerspace during the piloting phase of my data collection as well as the support from the many individuals from various hackerspaces in Denmark, Germany, Austria and the United States who shared their insights on hacker culture and beyond, but will remain unnamed to respect their anonymity. A special thanks however goes to Fin for his collaboration on the research design and for facilitating the access to the hacker communities. I am very grateful to the faculty, staff and my PhD fellows of the Department of Innovation and Organizational Economics for a tremendous education and research environment. I would like to express a special thank you to Professor Lee Davis for supporting my work throughout the years. Her advice in the data collection phase and also during the job market period was very precious. I have been very fortunate to pursue my research in this department and additionally, to get to know several wonderful supporters outside my home institution. I would like to thank Professor Howard Aldrich who kindly provided extremely helpful feedback on my PhD. He also encouraged me to look into new, highly interesting research questions, explore my data in creative ways and connected me with great researchers with a joint interest in the hacker and maker movement. The work of Professor Marc Gruber has inspired many parts of my PhD thesis and he has been very generous with his time and advice. His comments on the survey and papers were precious and if not pursued, the mistake is mine. He also helped me in my professional socialization and introduced me to scholars within the discipline at various conferences. I highly appreciate that. I would like to thank Professor Mitchell Stevens and Annette Eldredge for their hospitality during my visiting period at Scancor at Stanford University in 2012/2013. I benefitted a lot from the vibrant research environment at Stanford and the opportunity to expose my work and engage in scholarly discussions with junior and senior faculty across campus. iii I am particularly grateful to Professor Woody Powell for the insightful discussions in- and outside the seminar of Organizational Theory, but most importantly for his unflagging encouragement of my work and for being a wonderful promoter. I want to thank Professor Eric von Hippel, for encouraging me to pursue a PhD in the first place and for introducing me to the academic world. I am very grateful for his trust in my abilities and steady support. This clearly had a big stake in my academic career. I also want to thank Dr. Joseph Reger for his mentorship during my time at Fujitsu Technology Solutions and beyond. I benefitted tremendously from these years in the private sector. His inspiring leadership, advocacy for innovation and endorsement of my PhD significantly shaped my career. I want to thank the good friends that I have made throughout my time as a PhD student and with whom I will have life-long friendships. Gouya, thank you so much for your wonderful moral support and your cordial friendship. Erika, thank you for your contagious vitality and good spirit. Virgilio, my running buddy, PhD fellow and dear friend: Thank you so much for your immeasurable help and friendship in all these years and for always having a sympathetic ear for the bigger and minor things in life. I owe a special debt to Francesca who helped me immensely in the last year of my PhD. You are a great friend and researcher - and a brimful source of optimism. Virgilio and Francesca, I will miss you – and the Italian dinners, the simple cake and of course Halifax. I also want to especially thank my very good and lifelong friends back home in Bavaria for keeping strong bonds despite the distance: Danke, Regensburger Mädls, Danke Mädls von daheim! I most want to say a sincere thank you to Rosalie, Josef, Florian and Josef Halbinger for their unconditional support. My family has always been my tower of strength. Ich bin Euch unsagbar dankbar für Eure unendliche Unterstützung! Ihr seid’s immer für mich da und mein Fels in der Brandung. Ohne Euch hätte ich das nicht geschafft. To express my heartfelt gratitude I dedicate this dissertation to my parents. iv v ENGLISH SUMMARY New ventures are central to an economy’s welfare and substantial promoters of technological change and innovation. Extant research has identified individuals and their role in entrepreneurial processes as the core pillars of entrepreneurship. This PhD dissertation aims to refine our understanding of the fundamental relationship between individuals and their entrepreneurial activities. In order to empirically test the hypotheses built, the analyses are based on a unique dataset on individuals in hacker- and makerspaces, i.e. open communities with physical workspaces where individuals with a common interest in technology, computing, science, art and hacking culture can meet, socialize and collaborate. Given that hackers and makers engage in various entrepreneurial activities, this empirical setting offers crucial benefits when analyzing entrepreneurial individuals. The thesis combines the literatures on psychology and entrepreneurship and consists of three essays. The first essay focuses on the early activities in the entrepreneurial process and depicts how the individual’s creativity as well as different forms of motivation influence different process activities including the discovery and exploitation of opportunities. The second essay consists of a longitudinal analysis and examines how the identity of the founder relates to firm exit. The third essay applies computational linguistic tools on haiku poems written by hackers to investigate how self-confidence, social awareness and social influence are related with entrepreneurial experience, defined by the number of times the individual has been involved in new firm establishments. Overall, the thesis aims to contribute to the field of entrepreneurship by providing empirical evidence on the individual-level factors that shape entrepreneurial activities. 4 In investigating entrepreneurial activities from a psychological perspective, I use the notion of entrepreneurship as a process that is based on two major components, individuals and opportunities. Opportunities generally are situations that indicate the possibility to make profit (Casson 1982, Shane 2012). However, entrepreneurial opportunities are particular in not being merely useful, i.e. optimizing existing goods or increasing efficiency of existing processes, but involving novelty (e.g. Kirzner, 1997; Shane & Venkataraman, 2000). The thesis adopts the framework in Shane and Venkataraman (2000) which describes entrepreneurship as a process where opportunities exist, and are discovered and exploited. Discovery includes a) exposure in the sense of being located in information corridors, i.e. where opportunities in form of complementary information pieces exist, and b) recognition, i.e. making connections between those information pieces to form opportunities with new means-end relationships. Exploitation refers to the exploitation of opportunities through firm foundation, i.e. exploitation of an opportunity in the organizational setup of a new firm, or other forms of implementation in markets and hierarchies, for instance in the form of patents and trademarks (see figure 1 for an overview). In building on this body of entrepreneurship literature, I follow the well-established notion of entrepreneurship defined as a process, and contribute to current discussions on process models (McMullen & Shepherd, 2006; Alvarez & Barney, 2007; Alvarez et al., 2013). This thesis draws on Shane and Venkataraman’s (2000) seminal model for four reasons. First, the model is widely accepted and considered to be a core concept in entrepreneurship research (Aldrich & Cliff, 2003). However, while Shane and Venkataraman call the major process components “stages”, this thesis refers to them as “activities”. This choice acknowledges the recent conception of activities in the process as not necessarily strategic or chronological (Shane, 2012). This thesis treats the activities 5 within the entrepreneurial process as distinct and independent and as activities that are not mutually exclusive. This leads to the second reason which concerns the characteristic that the activities can cooccur or be iterative. For instance, it is possible that individuals might pursue several activities simultaneously or exploit opportunities to which they have not been exposed or discovered. The third reason refers to the fact that since information is unevenly distributed (Hayek 1945), not all opportunities are identified, recognized and exploited (Schumpeter, 1934; Casson, 1982; Kirzner, 1997; Shane & Venkataraman, 2000). Consequently, by treating each activity separately in a standalone way, the thesis takes account of this and acknowledges that individuals can pursue entrepreneurial activities independent of each other, and opt out at different points in the process. Fourth, entering information corridors in which opportunities arise (as noted earlier), thus also refers to an activity. This thesis deviates from Shane and Venkataraman’s model in terming this activity exposure. That is, the individual does (or does not) expose him- or herself to situations where opportunity-related information exists. The reason for this adaptation is to allow a coherent view and consistent terminology in relation to entrepreneurial activities and the role of the individual in the process. However, it is important to note that the underlying theoretical principle as proposed by the authors, remains unchanged, and thus, the term exposure is in line with Shane and Venkataraman’s process model. With respect to the overall thesis, the investigation does not stop with opportunity implementation or becoming an entrepreneur. Instead, in line with prior work, exit is included in the broader conception of the process (e.g. Wennberg, 2009) because, as noted earlier, the technological, regional and economic benefits only occur if newly founded organizations survive 6 (Bruederl et al., 1992). While there are various definitions and conceptions of exit, in this thesis, firm exit refers to performance, and is defined as firm discontinuance, i.e. the termination of business (Yang & Aldrich, 2012). Furthermore, since the entrepreneurial process does not have to be rational, strategic or chronological (Shane, 2012), in this thesis it is considered that individuals can pursue parts of the entrepreneurial process repeatedly, thereby increasing the breadth of their experience. Analyzing the factors associated with entrepreneurial experience is important given their associated benefits such as increased firm performance (e.g. Stuart & Abetti, 1990; Delmar & Shane, 2006; Dencker et al., 2009; Eesley & Roberts, 2012) and above-average economic growth (Plehn-Dujowich, 2010; Roberts & Eesley, 2011). In line with prior work, the thesis captures entrepreneurial experience by the number of examples of entrepreneurial activity, i.e. number of firm foundations (Stuart & Abetti 1990; Delmar & Shane, 2006; Hmieleski & Baron, 2009; Eesley & Roberts, 2012). Combining the psychology and entrepreneurship literatures is advantageous for several reasons. Overall, it facilitates an individual-level perspective on the notion of entrepreneurship as a process and extends work on the micro-foundations of entrepreneurship. More specifically, since the thesis combines three studies each of which independently represents a different part (subprocess) of the entrepreneurial process, the thesis provides detailed insights into the factors that are important for different entrepreneurial activities, i.e. activities before and including the transition to entrepreneurship (chapter 2), firm exit (chapter 3) and the extent of entrepreneurial experience captured by the number of firm foundations (chapter 4). From a theoretical point of view, this combination advances the knowledge in this field by providing insights into why there are more opportunities than entrepreneurs (chapter 2), why some 7 individuals repeatedly engage in entrepreneurial activities (chapter 4) and how entrepreneurs influence the rates of exit of their firms (chapter 3). Specifically, I show in chapter 2 that not all opportunities are exploited because there are different individual factors that matter for the early, pre-founding activities of entrepreneurship compared to those that matter for the later exploitation activities. Additionally, I show in chapter 3 that entrepreneurs pursue activities in a way that supports their self-conception thereby influencing their firms’ viability, and in chapter 4 I explain how factors that are important in interactions with others relate to the repetition of entrepreneurial activities. From an empirical point of view this combination is advantageous because first, it extends the literature on entrepreneurial process models by operationalizing distinct entrepreneurial activities as separate dependent variables and empirically testing Shane and Venkataraman’s seminal model (chapter 2). Second, chapters 3 and 4 particularly benefit from this combination because it allows the application of well-established constructs in psychology to the field of entrepreneurship, and operationalization of theoretical concepts at the individual-level through the use of novel empirical strategies to generate the explanatory and control variables. The thesis chapters connect in the sense that all the explanatory variables stem from the field of psychology and all the dependent variables originate from entrepreneurship research. Figure 1 shows that the dependent variables in each chapter represent distinct entrepreneurial activities which link the chapters in this thesis. ------------------------------------------ Insert Figure 1 about here ------------------------------------------- 8 Thesis Goal and Research Questions The overall goal of this thesis research is to substantiate the importance of the individual in entrepreneurship. Specifically, it examines the social psychological factors with respect to entrepreneurial process activities to provide empirical evidence on: a) how creativity as well as intrinsic and extrinsic motivation influence opportunity exposure, recognition and exploitation; b) how founder identity is linked to firm exit; and c) how social skills are associated with entrepreneurial experience. To address these research questions requires an empirical context of entrepreneurs and comparable non-entrepreneurs, observed before and after the transition to entrepreneurship. This imposed a methodological challenge with respect to finding appropriate data, a challenge acknowledged but not fully addressed in the entrepreneurship literature (Shane & Khurana, 2003). Figure 2 provides a graphical overview of the thesis chapters and titles, and highlights the research questions, dependent variables investigated, and the data and empirical strategies applied. ------------------------------------------ Insert Figure 2 about here ------------------------------------------- The Empirical Context: Hacker- and Makerspaces I collected data on individuals listed as registered members of hackerspaces and makerspaces. A hackerspace is an open community that provides workspace for like-minded individuals to meet, socialize and work together. Hence, the term hackerspace can refer to the community as well as the physical workspace. Typically, hackerspaces are operated by their 9 communities, and offer members physical and virtual platforms for problem-solving, ideas exchange and collaboration, in areas such as technology – especially computers - and the sciences and arts. Hackerspaces, sometimes called hackspaces, makerspaces, makerlabs or hacklabs, are financed mainly by membership fees and usually equipped with a broad range of tools, machinery, hardware and materials including PCs, and computer components, hammers, saws, 3D printers and materials such as wood and metal parts, etc. Hence, they require physical spaces or work rooms which typically are located in garages, basements, warehouses, factory buildings, education or social centers. A typical example is Metalab, a 230 sqm non-profit hackerspace, located in a basement in the first district of Vienna, Austria. Metalab provides physical space to foster open and free information exchange and collaboration between technology-minded individuals, so-called digital artists, hobbyists and entrepreneurs, who share an interest in information technologies, new media, arts and hacking culture generally. The workspace includes a central meeting room equipped with tables, chairs and projectors, to facilitate group working, talks, presentations and workshops. Metalab also has a library, a social space, a kitchen, bathrooms and several small labs, for instance for photography, oscilloscopes, laser cutters and 3D printers - all equipped with high speed internet connection. Metalab was founded in 2006, inspired by German hackerspaces, and has developed to be one of the most influential hackerspaces in Europe, playing a crucial role in establishing other hackerspaces around the globe, for instance the NYC Resistor in New York, USA. Metalab appears to be a hotbed of entrepreneurship and has successful firm foundations including Mjam, Austria's largest internet food delivery company, and YEurope, Europe's first YCombinator inspired startup accelerator. Like many other hackerspaces, Metalab is run by a voluntary organizing team which 10 meets regularly to discuss hackerspace related issues such as renovation projects and investment. Metalab is financed primarily by membership fees in addition to corporate and government funds. Members of hackerspaces are known as hackers and makers. The term “hacker” originally had a positive connotation, describing sophisticated computer experts who were able to use and develop computer programs, software and hardware beyond their original purpose. It later was seen as a negative description, through media association with computer breakins for malicious purposes such as exploiting leaks in state government security systems. As a result, debate emerged about the term. The mainstream conception of “hacker” is still associated with a, to some extent negative or even criminal connotation. However, experts in the field and the communities see the term “hacker” as neutral, and refer to individuals who engage in criminal hacking and illegal activities as “crackers”. The term “maker” refers to individuals who not only alter existing products but generate completely new concepts, designs and projects using hand tools and machinery such as 3D printers or computer numerically controlled (CNC) machines (Lang, 2013; Aldrich et al., 2014). This thesis adopts the neutral perspective and refers to hackers and makers to describe individuals who use and enhance existing technologies, products, materials or any goods beyond their original purpose, and engage in the development of their own new projects, works, concepts and designs. Hence, activities in hacker- and makerspaces go beyond problem-solving, code writing and programming. They include new ideas, and the creation of artifacts, solutions, products and services that are often highly innovative and entrepreneurial. Hacker- and makerspaces usually provide diverse tools including hammers, saws, needles or more expensive, technical equipment including 3D printers and CNC machines to facilitate experimentation and prototyping of ideas. Hacker- and makerspace members can subscribe to distribution and mailing lists that are addressed 11 to the community allowing them to benefit from the “wisdom of the crowd”. Since free information exchange is one of the most important maxims in hacker and maker communities, the thresholds for entering these communities are low, and by subscribing to the mailing lists, individuals become registered members in the space. In return, they are embedded in a community where problems, needs and ideas can be discussed and probed virtually without physical presence in the workspace, for instance via wikis, blogs, email or RSS feed. Consequently, hacker- and makerspace members can be constantly exposed to the diverse problems, needs and ideas of their peers in the community. This virtual and physical collaboration, allows the involvement of technologies in the projects hackers and makers pursue but is not limited to that. The project nature is extensive and ranges across creative problem-solving by reusing and modifying software and hardware, experimenting with light, sound and text, 3D printing of component parts, tools and equipment, game development and generating new artifacts and inventions for instance robots, machinery and drones. Hackers and makers typically alter existing goods when they experience problems or dissatisfaction with existing market offerings, which may result in the development of completely new market offerings. Consequently, hacker and maker projects vary in terms of degree of usability and innovation. An example of an innovative hacking project that resulted in firm foundation is the opensource RepRap Project (replicating rapid prototype). The goal of the RepRap Project was to develop a 3D printer able to print its own components, thereby applying a particular manufacturing technique that produces material and components in layers. One of the project’s founders, in 2009 co-founded MakerBot Industries in New York, a firm that produces 3D printers which enable users to produce three-dimensional objects of virtually any shape based on a digital model. In 2011, the firm attracted venture capital investment of US$ 10 million. In 2013, MakerBot Industries was 12 acquired in a stock deal worth US$403 million and currently serves its customers as a distinct brand and subsidiary of Stratasys Incorporated. The individuals in hacker- and makerspaces share a set of common values such as freedom of speech, transparency, independence and learning, to enable creativity and collaboration (Coleman & Golub, 2008). Since these values are central to their selves, hackers tend to be less inclined to take up positions in hierarchically organized firms, even were the latter willing to hire them (Carlson, 2011). The majority of hackers are well-educated either as autodidacts or because they are pursuing studies at a university or college (see also Lakhani & Wolf, 2005). If active in the labor market however, hackers select into occupations such as free-lancers and jobs that offer personal freedom and flexibility, primarily in the information technology (IT) industry, or become entrepreneurs. In general, the projects hackers pursue are based on personal interest and thus are less formally organized and tend to follow the maxims of the hackerspace of shared resources, ideas and labor. The motives for becoming a hacker vary; some individuals join a hackerspace in order to open a business (profit or non-profit), others do so to create art, or because they enjoy the activity of hacking and being part of a social group (Lakhani & Wolf, 2005; Carlson, 2011). Given the nature of technology, and in particular internet-based projects, collaborations exist within and across spaces since information and knowledge can be shared even with geographically distant communities. Consequently, hacker- and makerspaces can be incubators of entrepreneurial activity and their members crucial contributors to innovation (von Hippel, 1986; Franke & Shah, 2003) and entrepreneurship (Autio et al., 2013). Hackers and makers engage in diverse entrepreneurial activities by exposing themselves to complementary need-based and solution-based information 13 (e.g. von Hippel, 1988), solving problems (e.g. Jeppesen & Lakhani, 2010), discovering ideas with business potential (e.g. Lilien et al., 2002), and becoming entrepreneurs (e.g. Shah & Tripsas, 2007). Data Collection and Assembling of Sample As insights into a context matter in order to understand the meanings of activities and practices (Brewer 2000, Barley & Kunda, 2001), I gathered in-depth information on the norms and behavior in hacker- and makerspaces before beginning my data collection, which is in line with the approach adopted in previous studies in similar empirical settings (e.g. Jeppesen & Frederiksen, 2006). The field work involved on site visits to hacker- and makerspaces and conferences across Europe, interviews with selected study subjects in all the regions investigated, and screening of relevant online data. Figure 3 shows the iterative process of field work combined with diverse test studies (paper based and online pilots) supplemented by a think aloud systematic (Forsyth & Lessler, 1991) which allowed the extrapolation of a survey format as an appropriate research design. Thus, the study design, the questions and their sequence were developed in line with the empirical context. The web-based format was chosen because the communication, problem-solving activities and collaboration in hacker- and makerspaces primarily use IT. All responses were anonymous and voluntary since the interviews showed that confidentiality was an important issue for the communities. To ensure reliability and validity, the survey was tested off- and online and piloted within several communities. 20 Hacker- and Makerspaces: A beneficial Setting for Entrepreneurship Research This context is particularly advantageous for addressing the research questions for various reasons. First, hacker- and makerspaces are an interesting context to apply an individual-level perspective to different entrepreneurial activities because space members are highly entrepreneurial and involved in various aspects of the entrepreneurial process. For instance, members of these communities alter existing market offerings based on their personal needs, develop new solutions, generate business ideas and start new firms. Consequently, it was possible to observe individual characteristics associated with a) the pre-founding stages, i.e. opportunity exposure and recognition, b) different forms of exploitation, i.e. commercialization via patents or firm foundation, and c) firm exit. Hence, the sample contains non-entrepreneurs, entrepreneurs, including those who subsequently exited, and serial entrepreneurs. From a methodological point of view, this is important for operationalizing the dependent variables. Second, hacker- and makerspaces represent a distinct community whose members pursue the same values and beliefs thereby lowering heterogeneity across participants. With reference to codified hacker ethics (Himanen et al., 2002), individuals are ideologically in line, share knowledge, collaborate and use similar codes and language. This is a major advantage of the setting because it makes study subjects reasonably comparable and reduces potential bias in the results. This enables measurement of the individual determinants of entrepreneurship since relatively similar individuals are observed among whom subsets become entrepreneurs, exit and have varying entrepreneurial experience. This leads to a third main advantage of the setting. Despite the commonalities within the community, individuals in this setting vary. In other words, the shared norms and language function as a baseline from which it is possible to observe variety and to disentangle the differences 21 attributed to the individual. This is particularly beneficial when analyzing differences across identities which inevitably are linked to values, beliefs and actions (Hogg & Terry, 2000) and language as a means to express identity (Pennebaker, 2011). To use individuals in hacker and makerspaces to analyze identities was motivated also by interviews and observations during preparation for the data collection. The findings from the field work indicate the existence of different identities, for instance individuals who modify existing products, software and services and develop new ideas to satisfy their own needs and those of others in the community, in contrast to individuals who become entrepreneurially active to harm competition and possibly crack security systems. However, the predictions in chapter 3 on founder identities and firm exit where inspired by the theory and tested using data on the existence of relevant identities. Analysis of individual attributes in similar settings has been done in prior research, and in the same vein, this thesis draws also on the context of hacker- and makerspaces to examine diverse forms of motivation and creativity (Roberts et al., 2006; Shah, 2006; Lakhani & Wolf, 2005). Fourth, hacker- and makerspaces are particularly appropriate to study innovation and entrepreneurship given their organization into virtual as well as physical platforms that enable information exchange, knowledge sharing and collaboration for their members. Innovation, which is often linked with entrepreneurship (Schumpeter, 1934), involves creativity and represents a social process that links individuals with diverse backgrounds, motivations, skills and vocabulary (Feldman, 2002). While the internet represents an important facilitator in this process for accessing and leveraging information, geography is important to overcome the limitations of virtual collaborations such as transfer of tacit knowledge and organizing resources required for the creation of knowledge (Feldman, 2002). Hacker- and makerspaces combine the advantages of both worlds 22 since projects, in particular technology-related projects, can be developed and discussed online and elaborated in the physical space if necessary. As already noted, the internet is the core communication medium within and across hacker- and makerspaces. Finally, choosing hacker-and makerspaces as the empirical context to study entrepreneurship is beneficial given the number of implications associated with the context and its description as the 2nd industrial revolution: According to Aldrich and colleagues (2014), hacker- and makerspaces are likely potential hotbeds of user innovation since they provide resources, equipment and tools relevant for generating entrepreneurial opportunities and business ideas, and potentially reduce failure rates because developments and prototypes can be tested early on in the process. Given the ideology of sharing and collaboration, the context is interesting to study in terms of the underlying motivations and strategies individuals in this context pursue when founding and managing their firms since this can influence the market dynamics and competition in respective industries. Thesis Chapters Chapter 2 provides a deeper understanding about how individuals come across entrepreneurial opportunities, how they recognize them as such, and transform them into business ideas. To depict a broader range of the entrepreneurial exploitation of these ideas, the study is not limited to firm creation and considers further implementation modes in markets and hierarchies which should be of interest to both scholars and practitioners. Since these issues are linked to both individuals’ motivations and cognitive skills, chapter 2 analyzes the influence of different motivations, i.e. intrinsic and extrinsic, as well as creativity. Intrinsic motivation refers to the individual’s inherent interest in an activity, while extrinsic 23 motivation relates to the stimulus associated with outcomes, for instance when individuals pursue an activity because they expect a particular result from it (Ryan & Deci, 2000). For the purposes of this study, creativity refers to a cognitive skill that is purposefully applied by the individual in the sense that the individual decides to deploy his or her creative skill in entrepreneurial activity (Sternberg & Lubart, 1996; Sternberg, 2006). The study aims to contribute to entrepreneurship research, in particular the literature on entrepreneurial process models, by empirically analyzing Shane and Venkataraman’s (2000) seminal process model. Specifically, it provides empirical evidence on the theoretical prediction that intrinsic and extrinsic motivation are important for different entrepreneurial activities (Amabile, 1997). By showing that intrinsic motivation matters more in early process activities and is detrimental at later process points, the findings suggest explanations for why there are fewer entrepreneurs than opportunities. Additionally, the results indicate that creativity plays a beneficial role in all entrepreneurial activities. Thus the study contributes specifically to entrepreneurship research in relation to creativity (e.g. Ward 2004, Audretsch & Belitski, 2013), by using a perspective of creativity that has not been empirically applied to investigate distinct process activities. Chapter 3 extends analysis of the entrepreneurial process and to my knowledge is the first empirical study to investigate how the identity of founders is linked to firm exit. It contributes to a better understanding of the values and meanings associated with entrepreneurial behavior and should add to the emerging literature that analyses entrepreneurship from a social identity perspective, i.e. how individuals categorize themselves compared to others (Tajfel, 1972; Tajfel & Turner, 1979). Chapter 3 discusses firm exit dependent on founders’ identity which is a reasonable 24 association to investigate since founders pursue firm strategies that inevitably are linked, specific to and aligned with their identities (Fauchart & Gruber, 2011) thereby impacting on the viability of their firms. Methodologically, I apply principal component factor analysis to generate three founder identity variables based on theories developed in prior work (Ashforth & Mael, 1989). Furthermore, to appropriately model the time perspective of firm exit, the empirical strategy involves thorough consideration of potential censoring issues when creating the required longitudinal dataset. I take account that firm exit is not random but conditioned on the transition to entrepreneurship and apply a two-stage Heckman (1979) specification including an instrument variable. The key findings of the study reveal that community oriented founders active in problemsolving, i.e. masterminds, are less likely to exit while the opposite applies to founders aiming to contribute to the world, i.e. missionaries. These findings are supported by several supplementary analyses and suggest that limiting the analysis to primarily recent firm foundations, mavericks, means reckless founders inclined to rivalry, are less likely to exit. The study’s findings support the prediction that community-oriented and missionary-prone identities are particularly dominant in shaping entrepreneurial action. The study intends to add to prior work on founder identity as grounded in social identity, by providing empirical evidence on how different founder identities have different impact on firm exit. Also Chapter 4 takes account of the importance of the social perspective associated with the individual. Pursuing entrepreneurship requires particular skills given that entrepreneurial activities and firm foundation are profoundly social activities (Whetten & Mackey, 2002). Thus, how individuals, more precisely potential entrepreneurs, communicate and interact with others, matters. 25 This is especially important because these activities need to be performed repeatedly in order to achieve entrepreneurial experience. Understanding what underlies entrepreneurial experience in turn is important given the extraordinary benefits that entrepreneurial experience brings in relation to increased firm performance of subsequent firms and disproportionate economic impact. These considerations motivated the analysis in chapter 4 which aims to provide insights into the relationship between so-called social skills i.e. skills relevant to human interaction, and entrepreneurial experience i.e. the number of firms founded by the entrepreneur. The study is co-authored with Toke Reichstein and, like the preceding chapters, draws on data obtained from the online survey to operationalize the study’s dependent variable. It draws also on the text corpus of haiku poems to generate the study’s explanatory variables. Specifically, we apply computational linguistics to identify patterns in the use of personal pronouns, and use them as proxies for different individual social skills. This indirect measure of the skills associated with the individual makes the study less prone to potential bias related to hitherto self-reported measures of social skills in the literature. This empirical strategy introduces a new methodological approach into the field of entrepreneurship and contributes to individual-level research. Language reveals information about individuals and is, similar to fingerprints, stable over time (Pennebaker & King, 1999). In analyzing personal pronouns, we address the social aspect of the study since pronouns are highly social (Tausczik & Pennebaker, 2010). The empirical analysis involves zero-inflated negative binomial regression to take account for over-dispersion in the data related to zeros representing individuals who never start a firm and the peculiarity that the individual’s decision to start the first firm may be distinctively different from the decision related to subsequent firm founding. 26 The main findings in this chapter suggest that entrepreneurial experience is greater at high levels of self-confidence and social awareness. Self-confidence refers to the individual’s belief in his or her own abilities (Chen et al., 1998; Simon et al., 2000), and social awareness captures the “degree of consciousness of and attention to the other” (McGinn & Croson, 2004, p. 334). In contrast to our prediction, the findings suggest entrepreneurial experience to be negatively associated with social influence, the extent to which someone is able to alter others’ attitudes or behavior in social interactions (Baron & Markman, 2000) proxied by usage of “we”. Our post hoc analysis sheds light on the acknowledged ambiguity of “we”-related pronouns, and supports the notion that pronouns are contextual (Pennebaker, 2011). More precisely, it indicates that in the context of hacker- and makerspaces, use of the first person plural pronoun refers to expression of a shared identity rather than the influential “we” used in political speeches. Given the importance of language in the field of entrepreneurship (Aldrich & Fiol, 1994; Lounsbury & Glynn, 2001), the study’s findings contribute to the field by revealing underlying attributes of entrepreneurial individuals through examining pronouns and support the notion of language analysis as a promising research trajectory in social science. 27 REFERENCES Aldrich, H.E., & Cliff, H. 2003. The pervasive effects of family on entrepreneurship: Toward a family embeddedness perspective. Journal of Business Venturing, 18: 573–596. Aldrich, H. E., & Fiol, C. M. 1994. Fools Rush In? The Institutional Context of Industry Creation. Academy of Management Review, 19: 645-670. Aldrich, H. E., & Kim, P.H. 2007. Small Worlds, Infinite Possibilities? How Social Networks Affect Entrepreneurial Team Formation and Search. Strategic Entrepreneurship Journal, 1: 147-165. Aldrich, H. E., Ferdinand, J.-P., Halbinger, M., Mollick, E., Greenberg, J., Shah, S., Gorbatai, A. 2014. The democratization of entrepreneurship? Hackers, makerspaces, and crowdfunding, Academy of Management Symposium, Philadelphia. Allport, F. H. 1920. The influence of the group upon association and thought. Journal of Experimental Psychology, 3:159-182. Alvarez, S.A., & Barney, J.B. 2007. Discovery and creation: Alternative theories of entrepreneurial action. Strategic Entrepreneurship Journal, 1:11–26. Alvarez, S. A., Barney, J. B., & Anderson, P. 2013. Forming and exploiting opportunities: The implications of discovery and creation processes for entrepreneurial and organizational research. Organization Science, 24: 301-317. Amabile, T. M. 1996. Creativity in context. Boulder: Westview Press. Amabile, T. M. 1997. Entrepreneurial creativity through motivational synergy. Journal of Creative Behavior, 31:18-26. Amoros, J.E. & Bosma, N. 2014. Global entrepreneurship monitor 2013 global report. Global Entrepreneurship Monitor. Ashforth, B. E., & Mael, F. A. 1989. Social identity theory and the organization. Academy of Management Review, 14: 20-39. Audretsch, D. B., & Belitski, M. 2013. The missing pillar: the creativity theory of knowledge spillover entrepreneurship. Small Business Economics, 41: 819-836. Autio, E., Dahlander, L., & Frederiksen, L. 2013. Information exposure, opportunity evaluation and entrepreneurial action: An empirical investigation of an online user community. Academy of Management Journal, 56: 1348-1371. 28 Bandura, A. 1986. Social foundations of thought and action. Prentice Hall.: Englewood Cliffs, NJ. Barley, S. R., & Kunda, G. 2001. Bringing work back in. Organization Science, 12: 76–95. Baron, R.A., 2006. Opportunity recognition as pattern recognition: how entrepreneurs “connect the dots” to identify new business opportunities. Academy of Management Perspectives, 20: 104-119. Baron, R. A., & Ensley, M.D. 2006. Opportunity recognition as the detection of meaningful patterns. Management Science, 52: 1331–1344. Baron, R. A., & Markman, G. D. 2000. Beyond social capital: How social skills can enhance entrepreneurs' success. The Academy of Management Executive, 14: 106-116. Bates, T. 1995. Self-employment entry across industry groups. Journal of Business Venturing, 10: 143-156. Brewer, J. D. 2000. Ethnography. Philadelphia: Open University Press. Bruederl, J., Preisendoerfer, P., & Ziegler, R. 1992. Survival chances of newly founded business organizations. American Sociological Review, 57: 227-242. Busenitz, L., & Barney J. 1997. Differences between entrepreneurs and managers in large organizations: biases and heuristics in strategic decision-making. Journal of Business Venturing, 12: 9-30. Cardon, M. S., Wincent, J., Singh, J., & Drnovsek, M. 2009. The nature and experience of entrepreneurial passion. Academy of Management Review, 34: 511–532. Carlson, J. 2011. Earth based hackerspaces, Chaos Communication Camp, Presentation, Berlin; http://prezi.com/9osaw2iws5jj/earth-based-hackerspaces/. Carroll, G. R., & Swaminathan, A. 2000. Why the microbrewery movement? Organizational dynamics of resource partitioning in the US brewing industry. American Journal of Sociology, 106: 715-762. Casson, M. 1982. The entrepreneur. Totowa, NJ: Barnes and Noble Books. Chen, C., Greene, P., & Crick, A. 1998. Does entrepreneurial self-efficacy distinguish entrepreneurs from managers? Journal of Business Venturing, 13: 295-316. Coleman, E. G., & Golub, A. 2008. Hacker practice Moral genres and the cultural articulation of liberalism. Anthropological Theory, 8: 255-277. 29 Coleman, E. G.2010. The hacker conference: A ritual condensation and celebration of a lifeworld. Anthropological Quarterly, 83: 47-72. Delmar, F., & Shane, S. 2006. Does experience matter? The effect of founding team experience on the survival and sales of newly founded ventures. Strategic Organization, 4: 215-247. Dencker, J. C., Gruber, M., & Shah, S. K. 2009. Pre-entry knowledge, learning, and the survival of new firms. Organization Science, 20: 516-537. Eesley, C. E., & Roberts, E. B. 2012. Are You Experienced or Are You Talented?: When Does Innate Talent versus Experience Explain Entrepreneurial Performance?. Strategic Entrepreneurship Journal, 6: 207-219. Fauchart, E., & Gruber, M. 2011. Darwinians, communitarians, and missionaries: The role of founder identity in entrepreneurship. Academy of Management Journal, 54: 935-957. Feldman, M. P. 2001. The entrepreneurial event revisited: Firm formation in a regional context. Industrial and Corporate Change, 10: 861-891. Feldman, M. P. 2002. The internet revolution and the geography of innovation. International Social Science Journal, 54: 47-56. Feldman, M. P., Francis, & J., Bercovitz, J. 2005. Creating a Cluster While Building a Firm: Entrepreneurs and the Formation of Industrial Clusters. Regional Studies, 39: 129–141. Forsyth, B.H., & Lessler, J. 1991. Cognitive laboratory methods: a taxonomy. In P. Biemer, Groves, R. Lyberg, L. Mathiowetz, N. & Sudman, S. (Eds.), Measurement errors in surveys: 393- 418. New York: John Wiley. Franke, N., & Shah, S. 2003. How communities support innovative activities: an exploration of assistance and sharing among end-users. Research Policy, 32: 157-178. Fritsch, M., & Mueller, P. 2008. The effect of new business formation on regional development over time: The case of Germany. Small Business Economics, 30: 15–29. Hayek, F. 1945. The use of knowledge in society. American Economic Review, 35: 519-530. Heckman, J.J. 1979. Sample selection bias as a specification error. Econometrica, 47: 153-161. Himanen, P., Torvalds, L., & Castells, M. 2002. The Hacker Ethic. Random House. Hmieleski, K. M., & Baron, R. A. 2009. Entrepreneurs' optimism and new venture performance: A social cognitive perspective. Academy of Management Journal, 52: 473-488. 36 FIGURE 3 Data Collection Steps and Overview of Data Set 37 Chapter 2 MOTIVATIONS, CREATIVITY AND ENTREPRENEURIAL ACTIVITIES by Maria Halbinger ABSTRACT This study applies a psychology lens on the notion of entrepreneurship as a process in which individuals discover and exploit opportunities, i.e. firm foundation and other implementation forms. It theorizes that the influence of creativity and different forms of motivation vary based on entrepreneurial activity. Drawing on a dataset of entrepreneurially active individuals from hacker- and makerspaces across the globe, the findings show a positive relationship between intrinsic motivation and opportunity exposure and a negative relationship with becoming an entrepreneur; in contrast, extrinsic motivation is positively associated with opportunity exploitation (becoming an entrepreneur and opportunity implementation in other forms). The positive effects of creativity across all entrepreneurial activities provide a more nuanced view of how a creative thinking style relates to entrepreneurship. The study aims to increase our understanding of why there are fewer entrepreneurs than opportunities. 38 INTRODUCTION Entrepreneurship is acknowledged to be a driver of regional development (Feldman, 2001) and economic growth (Sternberg & Wennekers, 2005). It is therefore puzzling that relatively little has been done to map the factors that shape entrepreneurial activities. Researchers, policy makers and practitioners all struggle with the conundrum of why there are fewer entrepreneurs than opportunities (e.g. Roberts, 1991; Shane, 2001). Entrepreneurship research primarily addresses this puzzle by focusing on factors related to the environment (e.g. Audretsch, 1995; Aldrich, 2000) and to technology (e.g. Christensen, 1997; Shane, 2001), and has increased our understanding of the conditions hampering or fostering entrepreneurship. However, relatively less attention has been paid to uncovering individual-level effects (Thornton, 1999; Shane et al., 2003). Person-centered studies have investigated individual characteristics (e.g. Busenitz & Barney, 1997; Zhao & Seibert, 2006; Shane et al., 2010) and demographics (Bates, 1995; Lazear, 2005) but these contributions mainly emphasize differences between entrepreneurs and non-entrepreneurs. Some works do consider entrepreneurship as a process and investigate the transition to entrepreneurship (e.g. Sørensen, 2007; Oezcan & Reichstein, 2009; Elfenbein et al., 2010) but there are no empirical studies that investigate the individual characteristics associated with engagement in different types of entrepreneurial activities. This chapter tries to provide a more detailed understanding of the association between individual characteristics and entrepreneurial venturing. Prior research shows that entrepreneurship involves multiple activities that are tightly connected from an individual point of view but not mutually dependent. Entrepreneurship involves being exposed to, recognizing and implementing opportunities and establishing a firm (Shane & Venkataraman, 2000). We need to understand how individual factors are related to these different 39 activities. While researchers have speculated about the cognitive mechanisms important for the discovery of opportunities (e.g. Baron, 2006), and considered whether specific motivations are linked to the exploitation of opportunities (e.g. Shane et al., 2003), empirical work investigating how individuals’ creativity and motivations relate to these distinct entrepreneurial activities is scarce. This is partly due to the difficulty of collecting data from comparable individuals engaged in different process activities that include pre-founding activities and different exploitation forms. As a result, empirical work on creativity and motivation is mostly confined to experimental and management research settings where entrepreneurial activities can be observed before their exploitation, and measurement of creativity can be performed by peers or supervisors (e.g. Amabile, 1996; Tierney et al., 1999; Zhou and George, 2001; Grant and Berry, 2012). The entrepreneurial processes that occur outside such settings are mostly unexplored, and the few studies that discuss creativity and motivation are mainly theoretical (e.g. Amabile, 1997a; Shane et al., 2003) or focus on a single aspect of the entrepreneurial process, i.e. the transition to entrepreneurship and the incentives associated with being an entrepreneur (e.g. Lazear, 2005; Zhao and Seibert, 2006; Taylor, 1996). This strand of research generally ignores the distinct characteristics which cause individuals to opt out at certain points in the process. The introduction of a psychological perspective into entrepreneurship research is advantageous. It provides a useful framework for the task of identifying the associations between individual factors and single entrepreneurial activities, responding to calls for a more detailed analysis of entrepreneurship that takes account of the various exploitation of opportunities not just the organizational setup related to new venture creation (Shane et al., 2003; Shane, 2012). A psychology lens allows investigation of the intra-psychic dimensions of individuals, i.e. how they think and process information, which is important with respect to accessing and processing 40 opportunity-related information. It also allows examination of entrepreneurial action by revealing the underlying human behavior motivations. It is important to know more about these associations. The influence of creativity has been long debated (e.g. Schumpeter, 1934; Ward, 2004) and while researchers have found the creativity embodied in creative classes, to be crucial for regional development and higher economic returns (e.g. Florida, 2005; Lorenzen & Andersen, 2009), little is known about how individual creativity influences activities ex ante venturing. Entrepreneurial activity is based on specific motivation (Shane et al., 2003) and it has been theorized that these motivations vary according to the activity (Amabile, 1997a). Hence, entrepreneurship research would benefit from analysis of how creativity and different forms of motivation impact on the single activities of entrepreneurship. Specifically, in this study I examine the influence of creativity, intrinsic motivation, i.e. when an individual expends effort on a task for reasons of pure interest and learning, as opposed to extrinsic motivation, i.e. when individuals expend effort for a particular expected outcome (Ryan & Deci, 2000), on exposure, recognition and successful implementation of opportunities. The present research investigates whether intrinsic and extrinsic motivations as well as creativity are associated with different activities in the entrepreneurial process. Intrinsically motivated individuals are argued to expose themselves to opportunities through their search for new information inputs which positions them into information corridors that allow access to rich, opportunity-related information. In contrast, intrinsic motivation considers the founding event to be time-consuming, and the associated uncertainty brings feelings of low levels of control and a mismatch between personal skills and the occupation. Creativity facilitates the recombination of complementary pieces of information which trigger opportunity recognition. According to the notion in investment theory of creativity, to “buy low and sell high” (Sternberg, 2006, p. 87; 41 Sternberg and Lubart, 1996), creative individuals decide to use their creative skills to exploit opportunities and start a firm. Extrinsically motivated individuals tend to be results-directed, pursuing entrepreneurial activities in the expectation of payback or career progression, which makes them more likely to implement opportunities and select into entrepreneurship. I utilize data on hackers and makers because individuals in hacker- and makerspaces provide a good setting to investigate these issues. Hackers and makers are highly entrepreneurial which allows analysis of all entrepreneurial process activities including those before transition to entrepreneurship. Prior research suggests the presence of diverse forms or motivation and creativity (Lakhani & Wolf, 2005; Roberts et al., 2006; Shah, 2006; Alexy & Leitner, 2011); I combine theories of entrepreneurial activities and psychology and apply them to this context to examine individual-level factors in entrepreneurship. The reasonably closed hacker- and makerspace environment lowers heterogeneity and facilitates investigation of comparable individuals categorized as entrepreneurs and non-entrepreneurs. The study in this chapter aims to contribute primarily to the literature on entrepreneurial process models through the introduction of well-established psychological constructs into the field of entrepreneurship. The findings suggest that intrinsic motivation is beneficial at the front end of the process but detrimental in the back end, particularly for the chances of becoming an entrepreneur. Extrinsically motivated individuals are more likely to become entrepreneurs and to exploit opportunities in various implementation formats, such as patents and trademarks. Hence, this study increases understanding of the ambiguous nature of intrinsic and extrinsic motivation and proposes explanations for why opportunities may remain unexplored or unexploited. This has some interesting implications for research on team formation and search, since it provides insights into the personal characteristic that matter at different points in the process, and offers guidance for 42 organization in the search for business opportunities. The insights into the positive effects of creativity across all entrepreneurial activities, provides a more nuanced understanding of how this factor benefits entrepreneurship. The next section outlines the theoretical framework and hypotheses. The third section discusses specification of the sample and the data collection process, and presents the econometric approach and testing of hypotheses. The final section presents results, and concludes with a discussion of the research contributions, implications, limitations and opportunities for further work. ENTREPRENEURIAL PROCESS AND OPPORTUNITIES While entrepreneurship researchers, over the years, have developed different process models (e.g. Ardichvili et al., 2003; Baron, 2007), there is agreement that stimuli vary across activities (Autio et al., 2013), and that intrinsic and extrinsic motivation for instance, matter for different process activities (Amabile, 1997a). This article exploits the idea behind Shane and Venkataraman’s model (2000) that individuals pursue opportunities via three main activities: opportunity exposure, recognition and exploitation. The first activity, opportunity exposure, requires the individual to a) possess prior information that is b) complementary with the new information inputs. The second activity requires the individual to recombine heterogeneous information pieces thereby creating new means-ends relationships. The individual’s recognition of an opportunity involves recognizing the entrepreneurial value of the information. These two activities represent discovery of an opportunity (figure 1) but are distinct activities. The existence of opportunity-related information, and exposure to opportunity-related information are distinctively different from the ability to recombine the 43 information and recognize the value of an opportunity as it takes shape in the entrepreneur’s mind (Baron, 2006; Shane, 2012). The final activity is exploitation of an opportunity through a) new firm foundation or b) via other implementation forms within and outside existing organizations such as new products, services, patents, or trademarks. Firm foundation refers to the “institutional arrangement” of firm creation as one way to exploit an opportunity (Shane, 2012, p.13). In response to criticisms related to limiting entrepreneurship only to firm foundation (Shane & Venkataraman, 2000), this chapter takes account of other arrangements such as the implementation of opportunities within existing organizations or markets. The analysis applies Shane and Venkataraman’s process perspective on entrepreneurship, but highlights four aspects. First, in order to take into account that the process does not necessarily follow a sequential or strategic order (stages), which is an important aspect of the model (Shane, 2012), the present study refers to the process modules as “activities” rather than “stages”. This acknowledges the fact that activities can be pursued independently and are not mutually exclusive, and that an individual can exploit an opportunity that someone else has recognized. Second, individuals can perform activities iteratively or simultaneously since exposure and recognition are closely intertwined. Third, this notion allows the possibility that individuals may stop at different points in the process. This is consistent with the statement that information is unevenly distributed (Hayek, 1945) and thus, not all opportunities are recognized and exploited (Schumpeter, 1934; Casson, 1982; Kirzner, 1997; Shane & Venkataraman, 2000). Fourth, in line with Shane and Venkataraman’s process model, the process is theoretically initiated by the individual’s position in the information corridor. I call this activity “opportunity exposure” since individuals must expose themselves to situations where opportunity-related information exists. Taken together, the 44 entrepreneurial process provides a blueprint with individuals and opportunities at its core, in which the ideal-typical individual is the agent who progresses through the process activities through the pursuit of an entrepreneurial opportunity. In line with prior work, I define opportunities as situations where the possibility exists that products, services, models or process are introduced to a market with a potentially positive pricecost relationship (Casson, 1982; Shane, 2012). In previous studies, entrepreneurial opportunities are explicitly associated with new products and services rather than optimization or efficiencyincreasing mechanisms applied to existing products, services, models or processes. Although both types of opportunities are appropriate for certain contexts and uses, the entrepreneurial opportunity is inherently novel (e.g. Kirzner, 1997; Shane & Venkataraman, 2000). But due to the imbalance between opportunity-related information and people’s beliefs, not all opportunities that exist are identified, recognized and exploited entrepreneurially (Schumpeter, 1934; Hayek, 1945; Casson, 1982; Kirzner, 1997; Shane & Venkataraman, 2000). By implication, only some individuals become aware of existing opportunities, and only some individuals recognize their value and exploit these opportunities for entrepreneurial purposes. Hence, entrepreneurial behavior is affected by opportunities but they do not define the whole process (Shane et al., 2003). This chapter examines the individual-related rather than the opportunity-related aspects of the phenomenon, and proposes a distinct view on the key personal attributes related to opportunity discovery and exploitation. ------------------------------------------ Insert Figure 1 about here ------------------------------------------- 45 Sub-activities of Opportunity Discovery: Opportunity Exposure and Opportunity Recognition Opportunity discovery requires individuals a) to be exposed to information that complements their prior information and b) to successfully combine and leverage relevant information inputs necessary to recognize the opportunity’s value. The succeeding parts of this chapter discuss how different forms of motivation and creativity determine entrepreneurial activities. Motivation comprises two factors, intrinsic and extrinsic motivation, which are not mutually exclusive because an individual can perform a task that involves both motivation forms simultaneously (Amabile, 1983). The role of intrinsic motivation on opportunity exposure and entrepreneurship Intrinsic motivation refers to the individual’s inherent interest in and appreciation of a task or activity (e.g. Ryan & Deci, 2000; Amabile, 1996). Individuals that are highly intrinsically motivated primarily obtain satisfaction from pursuing particular activities without expectation of outcomes because they support their feelings of competence and self-determination (Deci & Ryan, 1985). For instance, research shows that intrinsic motivation underlies software developments because individuals are willing to dedicate time and effort to problem-solving and development activities for the sheer pleasure of doing so (Lakhani & Wolf, 2005; Roberts et al., 2006). Although not all individuals are merely intrinsically motivated in the moment of performing a particular task (Amabile, 1983), for instance in open software development (Alexy & Leitner, 2011), they can become drawn into the task if there is a perfect match between skills and the task which results in a “flow”, a state of mind without a sense of time, place or pressure which induces positive emotions (Csikszentmihalyi, 1996). This attracts individuals into situations where there is a wide range of information which extends their attention scope thereby increasing the possibility to receive new 52 hackers and makers as individuals who employ goods and services including technologies, products and materials of any kind, beyond their initial purpose, alter them and generate completely new work and projects including new concepts, designs and artifacts. Hackers and makers are usually members of hackerspaces, or hackerlabs and makerspaces or makerlabs. The terms hackerspace and makerspace is used to describe both the community of members and the physical workspace that the community provides for individuals with a shared interest in areas such as technology, computing, arts and science, to socialize, exchange ideas and collaborate. Hackerspaces can be located in education or social centers, industrial buildings, warehouses, garages or basements, and comprise labs equipped with a range of tools, materials and machinery including hammers, drilling machines, laser cutters, 3D printers and, sometimes specialized equipment, such as computer numerically controlled (CNC) machines and oscilloscopes. Hackerspaces are usually run by a community-internal organizing team and financed by (voluntary) membership fees, and corporate and public donations. The activities of hackers and makers involve more than solving problems related to existing products, and include programming and game development, and the creation of completely new works, designs and concepts such as robots, drones and art-related installations in combination with materials, sound, light and fire. These activities may be collaborative within and across hackerspaces, and hackers and makers collaborate both physically in the workspace, and virtually in the community. Since there is generally no bar to access, individuals can subscribe to hackerspace mailing lists to discuss problems and ideas related to hacking, which gives access to the knowledge and expertise of the community without the need for a physical meeting. Free information exchange, and resource and knowledge sharing represent core values of the hacker community (Coleman & Golub, 2008), and 53 these mailing lists are used frequently thereby constantly exposing its members to the needs, problems and solutions of its members. Hacker- and Makerspaces as an Appropriate Study Choice Studying individuals in hacker- and makerspaces had several motivations. First, studying hackers and makers to analyze the individual-related aspects of the entrepreneurial process is advantageous because their hacking activities refer to entrepreneurial activities. Research shows that these individuals encounter need-based and solution-based information (e.g. von Hippel, 1988), engage in problem-solving (e.g. Jeppesen & Lakhani, 2010), develop ideas with high potential commercial value (e.g. Lilien et al., 2002) and transition to entrepreneurship (e.g. Shah & Tripsas, 2007). Consequently, these individuals are crucial stakeholders in the processes of innovation (von Hippel, 1986; Franke & Shah, 2003) and entrepreneurship (Autio et al., 2013). Hackers and makers are interesting as study subjects to investigate the influencing factors of distinct entrepreneurial activities. Second, since firm foundation is the result of human action rather than environmental effects, macro level studies analyzing how context effects shape the individual’s likelihood to become an entrepreneur (e.g. Aldrich & Zimmer, 1986; Carroll & Swaminathan, 2000; Sorenson & Audia, 2000; Oezcan & Reichstein, 2009) have been criticized as ignoring individual-level effects (e.g. Thornton, 1999; Shane et al., 2003). To lower heterogeneity and disentangle the effects of individual characteristics, requires a setting with a reasonably homogeneous population. Apart from experimental studies, meeting this requirement is complex, and appropriate settings in natural environments are rare. Hacker- and makerspaces are ideal in this respect; they represent a closed community in which the main values 54 and ethics are shared (Himanen et al., 2002, Coleman & Golub, 2008). Although communities are open to new membership, typically only like-minded individuals join hacker- and makerspaces in order to collaborate with individuals with whom they have common interests. Prior work shows also that these communities represent contexts within which individuals dedicate enormous time and where members report their greatest creative achievements, e.g. in open software development (Lakhani & Wolf, 2005). Hence, this setting is appropriate also for analyzing diverse forms of motivation and creativity. The individual members of the community differ in their “mental equipment”, their capabilities, and how they exploit them (e.g. Jeppesen & Frederikson, 2006). Among those that discover an opportunity, only some become entrepreneurs (Shah & Tripsas, 2007; Mollick, 2012) while others prefer to commercialize their ideas (Jeppesen & Lakhani, 2010) or remain at stages prior to exploitation because of the enjoyment derived from entrepreneurial activity (Lakhani & Wolf, 2005). However, it should be noted, that such a setting could lead to a problematic population of study subjects in the sense that the variation in these individuals might not be reflected in a selected sample of potential respondents. For this reason, we include a control variable for area of development, region and relatedness between the hacking activity and the professional occupation in the empirical analysis. Data Collection and Assembling of Sample Data collection included (but were not limited to) preparatory field work and piloting phases which were performed iteratively. The insights into a setting are beneficial for increased understanding of the context’s activities, practices and related meanings (Brewer, 2000; Barley & 55 Kunda, 2001). I therefore conducted several field studies including web-analyses, and interviews and field observations in hackerspaces and at hacker conventions. The field work combined with several pilot studies including offline and online tests and a think-aloud systematic (Forsyth & Lessler, 1991), informed the research design and format of the online survey and the detailed individual and firm level questions. The data were collected stepwise, between May and July 2012. After the administrator of the biggest global hacker and maker consortium had promoted the survey online to signal its trustworthiness, the survey was distributed to 392 hacker- and makerspaces in Northern English- and German-speaking Europe 4 , the United States, Canada, Australia and New Zealand. Hackerspaces were selected based on criteria such as activity status (e.g. active versus closed) and purpose (e.g. private versus commercial purpose), membership conditions, and accessibility. I was unable to reach 23 hackerspaces, and 369 hackerspaces were sent reminders after eight to ten working days. By the end of the survey period, 2948 individuals had visited the survey platform. The final sample was extracted from that number by applying a response time window of 7.5 and 90 minutes, corresponding to the measured response timeframe in the pilot studies, which reduced the final sample to 678 individuals from 244 hackerspaces. While 43.8 percent of survey respondents in the final sample had started at least one venture, 76.6 percent of respondents were engaged in IT-related activities. The age span was 16-72 years with a mean age of 32.9 years old; 39.52 percent selfreported single as their marital status and 17.1 percent reported having children. The gender distribution was 57 female and 498 male individuals (with 123 missing values). 4 The European countries include Austria, Denmark, Finland, Germany, Norway, Sweden, Switzerland, U.K. and Ireland 56 Limitations of the Setting Despite the advantages of the setting noted earlier, the sample suffers from some shortcomings. First, the concerns over privacy occupying numerous hacker- and makerspace administrators resulted in constraints with respect to report response rates and representativeness checks of the sample 5 . Second, the final sample appears unbalanced with respect to the distribution of respondents across a) regions (105 from Europe, 94 from the United States and Canada, 21 from Australia and New Zealand and 24 from other countries) and b) hackerspaces since 30 out of 244 hackerspaces had five or more respondents. Although it is not possible to test formally to address these potential issues, several checks were performed to compare the final sample with studies in similar contexts in entrepreneurship and the respective literature, and my field observations. First, with respect to demographics, the sample’s over-representativeness of male respondents with 89.72 percent appears to reflect the standard in the field (e.g. Carlson, 2011; Coleman, 2010), and is consistent with the notion in research that entrepreneurship is a male domain (e.g. Hout & Rosen, 2000), e.g. on one study of entrepreneurship that adopts a social cognitive perspective there are 81.10% of male respondents (Hmieleski & Baron, 2009), and is in line with studies conducted in comparable contexts, i.e. problem-solving activities - 90.0% males (Jeppesen & Lakhani, 2010) and hacking - 97.5% males (Lakhani & Wolf, 2005). A check on age showed that the final sample with a mean age of 32.9 years is in line with other hacking studies that report mean ages of 30 years (Lakhani & Wolf, 2005) and 29 years (Jeppesen & Frederiksen, 2006), and that the age range of 90.37% of the respondents of 16-48 is 5 For instance, several hackerspaces did not reveal the number of registered members in the community. 57 well within the range considered for individuals to be prone to entrepreneurship (Aldrich & Kim, 2007; Oezcan & Reichstein, 2009). In addition, the high reporting of computing, technology, development of programs, applications and games, as main areas of respondents’ activities is consistent with field observations (e.g. Carlson, 2011), research on anthropology (Coleman, 2010) and studies in similar settings (Lakhani & Wolf, 2005). Second, to check for potential bias related to the regional unbalance I compared entrepreneurial activities across regions based on the Total early-stage Entrepreneurial Activity measure (TEA). This key measure, from the Global Entrepreneurship Monitor 2013 report (Amoros & Bosma, 2014), exhibits trends consistent with the present study’s final sample, in the sense that entrepreneurial activity is higher in the United States and Canada compared to Europe. 6 Australia and New Zealand are not included in this report but appear to have similar response and entrepreneurship rates at 8.61% and 8.08% respectively. I also controlled for region in the empirical analyses which indicate significant effects on the activity of firm foundation but do not indicate reasons for concern, since the main results are unchanged with the inclusion of the control. Third, although the tests on the demographic dimensions mentioned above indicate that the final sample is reasonably representative, I included in the econometrical analyses cluster corrected standard errors based on hackerspaces as the cluster identifier. The results did not change. Overall, the tests indicate that the final sample matches well with observations in the field and studies investigating similar research questions and contexts. 6 The TEA measure includes individuals between 16 and 64 who are about to establish or already run a business. 58 Measures Dependent variables This section discusses operationalization of the dependent variables for the activities involved in the entrepreneurial process, from opportunity exposure, recognition, to exploitation via firm foundation and other implementation formats. All the dependent variables are operationalized as dummy variables taking the value 1 in the case that the respective activity is pursued by the individual and 0 in case it is not pursued. This operationalization is in line with the notion of an entrepreneurial process in which activities can be carried out independently and not necessarily in sequential order. Opportunity exposure. The dependent variable is measured by a proxy capturing two relevant aspects of exposure: a) the individual’s prior knowledge that is complementary to b) new information inputs (Kaish & Gilad, 1991; Shane & Venkataraman, 2000). In line with previous research I operationalize the first component, prior knowledge, through the user’s information about needs (von Hippel, 1988) or unsolved problems (Venkataraman, 1997) and combine it with the second component, captured by a dummy variable indicating whether or not the individual encounters new pieces of information (Shane & Venkataraman, 2000). In the presence of both types of information available to the individual, heterogeneous pieces of information encounter each other. Thus, the dummy variable indicates whether or not the individual is exposed to an opportunity. The variable is generated stepwise by combining responses to the question “I hack because I have unmet need(s) or unsolved problem(s) I want to solve” with the number of hackerspace visits (online or physically), indicating that the individual has obtained new needs- and solutions-based information from other members. 59 Opportunity recognition. To measure opportunity recognition, I exploited a questionnaire item on the individual’s hacking and development history. The question took the style of a question on firms’ innovation developments in the U.K. innovation survey which is based on the Eurostat Community Innovation Survey (CIS) (DTI 2003). In contrast to the rather objective phenomenon of being exposed to an opportunity, this dependent variable captures the individual’s decision about what action to take upon exposure to opportunity-related information pieces. The combination of resources and information stocks incorporates the prerequisite that the individual must have recognized a new means-ends relationship. Thus, opportunity recognition is rather subjective and related only to the pure existence of an opportunity (Shane, 2012) and exposure to it. Based on this argument, I measure whether the individual has performed this combination task and generated a development based on his/her hacking activities. More precisely, with reference to this CIS survey type question I generated a dummy variable for whether or not the community member had made a significantly new development or significantly 7 improved a new technology, a new combination of existing technologies, or other knowledge, material or information. From this perspective, the information about developments based on novel combinations of heterogeneous information pieces, proxies for whether or not an opportunity has been recognized. Firm foundation. To construct the variable for firm foundation and transition to entrepreneurship, I generated a dummy for whether or not the individual had founded or co-founded a company. Opportunity implementation. In line with prior work, this study applies a less restrictive definition of entrepreneurship where exploitation includes both firm foundation and the implementation of opportunities in markets or hierarchies (e.g. Shane, 2012). For a refined analysis 7 “Significantly new” was thereby referring to the peers in the community for a reference point 60 of the determinants, the different forms of exploitation are measured separately as opportunity implementation and firm foundation. To construct opportunity implementation formats, I created a dummy variable based on survey items about number of new products, services, patents or trademarks realized as individual developments. Independent variables The independent variables are based on measurement items developed and used in social psychology. Unless otherwise indicated, they are used in the original scale version. Two separate principal component factor analyses were conducted involving the Kaiser criterion (Kaiser, 1960) and varimax rotation for the survey items related to motivation and creativity. The two factors required to operationalize the intrinsic and extrinsic motivation variable result from a factor analysis performed on all motivation-related survey questions. By construction, these two factors are not correlated and thus are not mutually exclusive, which is an important requirement according to the theory. An overview of the survey questions, factor loadings, variances and Cronbach’s alphas are summarized Table 1. Since the dataset involves categorical variables, alternative factor analyses were conducted. For this purpose, I performed a polychoric correlation matrix and exploratory factor analysis in which the matrix rather than the raw variables function as the input. This alternative technique delivered similar but weaker results which represents additional support for the principal component factor analyses results. ------------------------------------------ Insert Table 1 about here ------------------------------------------- 61 Intrinsic motivation. Operationalization of intrinsic motivation variable was performed in line with self-determination theory measuring different forms of motivation (Deci & Ryan, 1985; Ryan & Connell, 1989). I conducted a principal component factor analysis with all the survey items related to intrinsic and extrinsic motivation. For the particular context of hacker- and makerspaces, I adapted the items based on prior studies using similar settings where individuals engage in entrepreneurial activities before firm foundation (e.g. Lakhani & Wolf, 2005; Roberts et al., 2006), and also on the insights gathered during familiarization with the empirical context. Hence, the items “I enjoy the activity of hacking itself”, “I enjoy being part of a community”, and “I forget everything around me when I get into the Zone” functioned as proxies for intrinsic motivation. The last item refers particular to “flow”, a state where individuals enjoy an activity to the point that they lose any sense of time due to a perfect match of skills and task (Csikszentmihalyi, 1996; Lakhani & Wolf, 2005). The three items loaded into one factor as expected. Table 1 shows further related statistics. Creativity. Individual level creativity has been defined and measured in multiple ways across experiments and field studies, and rated in various formats involving peers, experts or supervisors (e.g. Amabile, 1979; Shalley & Perry-Smith, 2001; Grant & Berry, 2011). Table 1 shows my operationalization of creativity using an adapted short version of a four-item scale developed by Sternberg (1985a). The self-reported measure is based on a seven-point Likert scale from 1 “strongly disagree” to 7 “strongly agree”. I chose this specific self-reported measure due to the underlying theoretical view of creativity highlighted above. The study considers creativity from the perspective of the individual’s decision about how to deploy the creative skills available to him or her (Sternberg, 2006). Hence, the analysis investigates a psychological process within the individual’s mind, rather one involving individuals or the environment. Since this mental process is 68 disentangles exploitation related to the notion of firm foundation from exploitation for other purposes such as new products, services, patents and trademarks. Second, the findings also add to research on motivation. Prior work in this research stream has debated whether the relationship between intrinsic and extrinsic motivation is antagonistic or synergetic. While the former associates an increase in extrinsic motivation for an activity with a decrease in intrinsic motivation, the latter refers to a relationship in which intrinsic and extrinsic motivation are self supporting, or at least do not negatively affect one another (see e.g. Deci and colleagues (1999) for a meta-analytical review). Research has increased our knowledge about the nature of this relationship in experimental settings and work environments in established organizations (e.g. Amabile, 1993, 1997b) and highlighted that the relationship is particularly complex in distributed models of innovation such as open software development (Alexy & Leitner, 2011). The present study does not measure the mutual effects of these motivations but is nevertheless interesting since the findings suggest that, investigating the whole entrepreneurial process, it appears that intrinsic motivation is required to initiate the process and extrinsic motivation is required to complete it. However, this study treats these activities separately and the results indicate that intrinsic and extrinsic motivation have significant, but opposite effects on the likelihood of starting a firm or of exploitation in other formats. Thus, this study contributes by empirically confirming a hitherto theorized variation across the entrepreneurial process (Amabile, 1997a). On the other hand, in providing empirical evidence that intrinsic motivation exhibits opposite effects in the early versus late process activities, the results are consistent with one of the few empirical studies on intrinsic motivation in entrepreneurship (Gimeno et al., 1997), which suggests that when individuals are intrinsically motivated they benefit from psychic incomes and 69 less from (financial) payback thereby decreasing the survival chances of their firms. The present study complements this work by examining the transition to entrepreneurship and providing potential explanations for this phenomenon related to time-constraints, uncertainty and person-job mismatch. Third, the connection between creativity and entrepreneurship appears to be intuitive and has been thoroughly discussed in the literature (e.g. Schumpeter, 1934; Amabile, 1997a; Ward, 2004). Empirical evidence on this relationship however, is scarce, in particular with respect to single entrepreneurial activities. This may be due in part to lack of appropriate settings to measure creativity based on activities before the transition to entrepreneurship. The findings in this chapter on the significant impact of creativity on single activities contributes to work in cognitive science and pattern recognition in entrepreneurial and invention processes (e.g. Amabile, 1997; Baron, 2006; Baron & Ensley, 2006; Maggitti et al., 2012). The use of data on individuals that engage in diverse entrepreneurial activities introduced a new empirical setting to research on creativity, in particular investment theory of creativity (Sternberg & Lubart, 1996). In line with this notion, the results suggest that individuals who intentionally deploy their skills for creative purposes, connect complementary pieces of information required to see new means-ends relationships and recognize opportunities. Moreover, the findings indicate that these individuals do not stop at this stage but are more likely to go on to exploit the recognized opportunities via firm foundation or other activities. Apparently, once individuals have been exposed and made the decision to invest their skills, time and efforts to transform information pieces into business ideas, they desire a “payback”. Based on the principle “buy low and sell high” (Sternberg, 2006, p.87), they implement the recognized opportunity in the form of firm creation or other exploitation forms. 70 Fourth, the study offers a complementary perspective on the literature on founding teams. Studies in this research stream debate whether diversity in team formation is preferable to homogeneity (e.g., Pelled et al., 1999; Ruef et al., 2003). This study’s findings extend this notion and indicate that even before firm foundation, diverse individual-level factors are important. Specifically, separating entrepreneurial activities from one another and analyzing them independently, gives an idea of the individual attributes that matter in the single activities of the entrepreneurial process. In other words, this chapter should provide guidance with respect to the search for individuals supporting prospective entrepreneurs, in case they are not “fully equipped” with the relevant motivations and creativity to successfully perform different entrepreneurial activities, and the roles required. Finally, the findings have implications for the search literature, particularly distributed sources of innovation, which is a recent link in the literature (Alvarez et al., 2013). Studies in this stream argue that companies increasingly aim to innovate by boundary-spanning search for innovation inputs, stemming from distributed sources of innovation and their positive impact on innovation processes (e.g. Katila & Ahuja, 2002; Laursen & Salter, 2006). This study’s results complement this research by providing insights into the distribution of potential sources of innovation and the underlying mechanisms with respect to the availability of opportunities in the market. The findings suggest that if individuals are not equipped with relevant motivations and creativity, opportunities that are discovered might remain unexploited despite their commercial value and their potential value as inputs to the organization’s innovation processes. This is consistent with the statement that sophisticated users become entrepreneurs by chance, if no existing organization can be found to exploit the opportunity (Shah & Tripsas, 2007). These individuals’ ideas represent interesting potential sources of innovation and are of high economic 71 interest to organizations due to their significantly higher potential for innovation and market success (Lilien et al., 2002). The study’s findings contribute to the literature on idea management (e.g. Alexy et al., 2012 ) by shedding light on where and why business opportunities emerge – which might be beneficial for a further investigation of measures, tools and management processes in order to translate external ideas into company internal innovation. Implications for Practitioners and Educators This research should be useful to venture capitalists and innovation managers. When founding or innovation teams are formed, the combination of the “right” people, or more specifically the “right” motivations and creative skills, has been shown to be crucial. Individuals intrinsically motivated in a particular activity can be beneficial to start the process. But their impact can be counterproductive in later stages. The research in this chapter provides evidence about the determinants of entrepreneurial and innovation processes, and moreover, suggests ideas on when and how different team members should be composed. Furthermore, since organizations typically search for new ideas to fuel their innovation processes, managers could benefit from knowing more about when and why entrepreneurial opportunities are available in the market. This should be of particular interest to high-tech industries where innovation cycles are short, competition is high, and many products are standardized; opportunities that are already in the discovery stage represent highly valuable inputs to entrepreneurial or innovation pipelines. Finally, the study should be of interest to educators in suggesting creativity training for nascent entrepreneurs and prospective founding teams. One approach might be proactive training of individuals in how to use creativity tools and methods to stimulate a creative thinking style, and 72 abandonment of conventional paths in the search for new ideas and ways to solve problems. An appropriate teaching approach might include awakening individual awareness in the various effects of motivation. Limitations and Future Research The particular context of hacker and makerspaces offers advantages for an analysis of the impact of motivation and creativity on the entrepreneurial process, including pre-founding. However, it reduces generalizability of the findings. The individual’s affiliation with a hackerspace might be a product of predisposition. Hence, the occurrence of entrepreneurs in our sample might be attributable to self-selection mechanisms, first to the hackerspace in general, and also to responding to the survey. The specifics of the data collection design did not allow me to check the representativeness of the final sample or response bias. However, I conducted several checks associated with respondents’ gender, age, hackerspace affiliation and geographic location. Comparisons with the fieldwork as well as studies analyzing comparable research questions in entrepreneurship and context-related studies suggest good correspondence between the final sample and this work. Future work could extend the investigation to other areas, such as processes in R&D labs, to study the determinants of invention or organizational innovation processes. With respect to the significance of the region control on firm foundation, future work could investigate how hackers’ proclivity for firm foundation varies across regions. The respondents in our sample ranked their skills and preferences for ideas and combinatorial tasks; thus measurement of the creativity variable could be criticized. However, as an individual’s tendency for deploying of his or her creative skills underlies intra-psychic mechanisms, 73 a self-reported measure is an appropriate way to measure creativity under the premise of the investment theory of creativity (e.g. Sternberg, 2006). Future research opportunities could include indirect analysis of individuals’ characteristics and skills. It could adopt a more holistic approach to analyzing individuals’ entrepreneurial endeavors. Prior work suggests that intrinsically motivated individuals not only experience an interest and enjoyment in the task but do so through an act of self-expression (Amabile, 1997a). Future work could take account of this view and analyze how entrepreneurship relates to individuals’ self-conceptions and how they are associated with firm performance. 74 REFERENCES Aldrich, H. 2000. Organizations evolving. Beverly Hills: Sage. Aldrich, H. E., Fiol, C. M. 1994. Fools Rush In? The Institutional Context of Industry Creation. Academy of Management Review, 19: 645-670. Aldrich, H. E., & Kim, P.H. 2007. Small Worlds, Infinite Possibilities? How Social Networks Affect Entrepreneurial Team Formation and Search. Strategic Entrepreneurship Journal, 1: 147-165. Aldrich H.E., & Zimmer, C. 1986. Entrepreneurship through social networks. In D. Sexton & R. Smilor (Eds.), The art and science of entrepreneurship: 3-23. Cambridge, MA: Ballinger. Alexy, O., & Leitner, M. 2011. A fistful of dollars: Are financial rewards a suitable management practice for distributed models of innovation? European Management Review, 8: 165-185. Alexy, O., Criscuolo, P., & Salter, A. 2012. Managing Unsolicited Ideas for R&D. California Management Review, 54:116-139. Alvarez, S. A., Barney, J. B., & Anderson, P. 2013. Forming and exploiting opportunities: The implications of discovery and creation processes for entrepreneurial and organizational research. Organization Science, 24: 301-317. Alvarez, S., & Parker, S. 2009. Emerging firms and the allocation of control rights: A Bayesian approach. Academy of Management Review, 34: 209–227. Amabile, T. M. 1979. Effects of external evaluation on artistic creativity. Journal of Personality and Social Psychology, 37: 221–233. Amabile T. M. 1983. The social psychology of creativity. New York: Springer-Verlag. Amabile, T. M. 1996. Creativity in context. Boulder: Westview Press. Amabile, T. M. 1993. Motivational synergy: Toward new conceptualizations of intrinsic and extrinsic motivation in the workplace. Human Resource Management Review, 3: 185-201. Amabile, T. M. 1997a. Entrepreneurial creativity through motivational synergy. Journal of Creative Behavior, 31:18-26. Amabile, T. M. 1997b. Motivating creativity in organizations: on doing what you love and loving what you do. California Management Review, 40:39-58. Amabile, T.M., Barsade, S.G., Mueller, J.S., & Staw, B.M. 2005. Affect and creativity at work. Administrative Science Quarterly, 50:367-403. 75 Amoros, J.E. & Bosma, N. 2014. Global entrepreneurship monitor 2013 global report. Global Entrepreneurship Monitor. Ardichvili, A., Cardozo, R., & Ray, S. 2003. A theory of entrepreneurial opportunity identification and development. Journal of Business Venturing, 18:105-123. Audretsch, D. B. 1995. Innovation and Industry: Evolution. The MIT Press. Autio, E., Dahlander, L., & Frederiksen, L. 2013. Information exposure, opportunity evaluation and entrepreneurial action: An empirical investigation of an online user community. Academy of Management Journal, 56: 1348-1371. Bandura, A. 1986. Social foundations of thought and action. Prentice Hall.: Englewood Cliffs, NJ. Bandura, A. 1997. Self-efficacy: The exercise of self control. New York: Freeman. Barley, S. R., & Kunda, G. 2001. Bringing work back in. Organization Science, 12: 76–95. Baron, R.A., 2006. Opportunity recognition as pattern recognition: how entrepreneurs “connect the dots” to identify new business opportunities. Academy of Management Perspectives, 20: 104-119. Baron, R. A. 2007. Behavioral and cognitive factors in entrepreneurship: Entrepreneurs as the active element in new venture creation. Strategic Entrepreneurship Journal, 1: 167–182. Baron, R. A., & Ensley, M.D. 2006. Opportunity recognition as the detection of meaningful patterns. Management Science, 52: 1331–1344. Bates, T. 1995. Self-employment entry across industry groups. Journal of Business Venturing, 10: 143-156. Blanchflower, D., & Oswald, A. 1998. What Makes An Entrepreneur? Journal of Labor Economics, 16: 26-60. Brewer, J. D. 2000. Ethnography. Philadelphia: Open University Press. Busenitz, L., & Barney J. 1997. Differences between entrepreneurs and managers in large organizations: biases and heuristics in strategic decision-making. Journal of Business Venturing, 12: 9-30. Campbell, J.P. & Pritchard, R.D. 1976. Motivation theory in industrial and organizational psychology. In M.D. Dunnette (Ed.), Handbook of Industrial and Organizational Psychology. Chicago: Rand McNally. 76 Campbell, C.A. 1992. A decision theory model for entrepreneurial acts. Entrepreneurship Theory and Practice, 17: 21–27. Carroll, G. R., & Swaminathan, A. 2000. Why the microbrewery movement? Organizational dynamics of resource partitioning in the US brewing industry. American Journal of Sociology, 106: 715-762. Carlson, J. 2011. Earth based hackerspaces, Chaos Communication Camp, Presentation, Berlin; http://prezi.com/9osaw2iws5jj/earth-based-hackerspaces/. Casson, M. 1982. The entrepreneur. Totowa, NJ: Barnes and Noble Books. Chen, C., Greene, P., & Crick, A. 1998. Does entrepreneurial self-efficacy distinguish entrepreneurs from managers? Journal of Business Venturing, 13: 295-316. Christensen, C. 1997. The innovators dilemma. Cambridge: Harvard Business School Press. Coleman, E. G., & Golub, A. 2008. Hacker practice Moral genres and the cultural articulation of liberalism. Anthropological Theory, 8: 255-277. Coleman, E. G.2010. The hacker conference: A ritual condensation and celebration of a lifeworld. Anthropological Quarterly, 83: 47-72. Collins, C., Locke, E., & Hanges, P. 2000. The relationship of need for achievement to entrepreneurial behavior: a meta-analysis. Human Performance, 17: 95–117. Csikszentmihalyi, M. 1996. Creativity: Flow and the psychology of discovery and invention. New York: Harper Collins. Davidsson, P. 2004. Researching entrepreneurship. New York: Springer. Deci, E.L., & Ryan, M.R. 1985. Intrinsic motivation and self-determination in human behavior. New York: Plenum Press. Deci, E. L., Koestner, R., & Ryan, R. M. 1999. A meta-analytic review of experiments examining the effects of extrinsic rewards on intrinsic motivation. Psychological Bulletin, 125: 627- 668. Donnellan, M.B., Oswald, F.L., Baird, B.M., & Lucas, R.E. 2006. The mini-IPIP scales: Tiny-yet- effective measures of the Big Five factors of personality. Psychological Assessment, 18: 192–203. DTI. 2003. U.K. Innovation Survey. Department of Trade and Industry, London (www.dti.gov.uk). October 2010. 77 Dunn, T., & Holtz-Eakin, D. 2000. Financial capital, human capital, and the transition to selfemployment: Evidence from intergenerational links. Journal of Labor Economics, 18: 282– 305. Dunne, T., Roberts, M., & Samuelson, L. 1988. Patterns of firm entry and exit in U.S. manufacturing industries. Rand Journal of Economics, 19: 495-515. Elfenbein, D. W., Hamilton, B. H., & Zenger, T. R.2010. The small firm effect and the entrepreneurial spawning of scientists and engineers. Management Science, 56: 659-681. Estrada, C. A., Isen, A. M., & Young, M. J. 1997. Positive affect facilitates integration of information and decreases anchoring in reasoning among physicians. Organizational Behavior and Human Decision Processes, 72: 117-135. Feldman, M. P. 2001. The entrepreneurial event revisited: Firm formation in a regional context. Industrial and Corporate Change, 10: 861-891. Florida, R. 2005. Cities and the creative class. Routledge. Forsyth, B.H., Lessler, J. 1991. Cognitive laboratory methods: a taxonomy. In P. Biemer, Groves, R. Lyberg, L. Mathiowetz, N. & Sudman, S. (Eds.), Measurement errors in surveys: 393- 418. New York: John Wiley. Franke, N., & Shah, S. 2003. How communities support innovative activities: an exploration of assistance and sharing among end-users. Research Policy, 32: 157-178. Fredrickson, B. L. 2001. The role of positive emotions in positive psychology: The broaden-and- build theory of positive emotions. American Psychologist, 56:218-226. Gagné, M., & Deci, E.L. 2005. Self-determination theory and work motivation. Journal of Organizational Behavior, 26: 331–362. Gimeno, J., Folta, T.B., Cooper, A.C., Woo C.Y. 1997. Survival of the fittest? Entrepreneurial human capital and the persistence of underperforming firms. Administrative Science Quarterly, 42: 750–783. Grant, A.M., & Berry, J.W. 2012. The necessity of others is the mother of invention: Intrinsic and prosocial motivations, perspective taking, and creativity. Academy of Management Journal, 54: 73-96. Hayek, F. 1945. The use of knowledge in society. American Economic Review, 35: 519-530. Himanen, P., Torvalds, L., & Castells, M. 2002. The Hacker Ethic. Random House. Hmieleski, K. M., & Baron, R. A. 2009. Entrepreneurs' optimism and new venture performance: A social cognitive perspective. Academy of Management Journal, 52: 473-488. 84 TABLE 1 Motivation and Creativity: Survey Questions, Factor Loadings, Variances and Cronbach’s Alphas Motivation Factor Analysis Questions Intrinsic Extrinsic I hack because... ...I enjoy the activity of hacking itself 0.812 ...I enjoy being part of a community 0.664 ...I forget everything around me when I get into the Zone 0.635 ...I would like to discover a business opportunity 0.849 ...I want to enhance my career opportunities 0.801 ...the hacker community gives support to found a company 0.758 Variance 1.515 1.959 Proportion 0.253 0.327 Cumulative 0.579 0.327 Cronbach's Alpha 0.478 0.730 Creativity Factor Analysis Questions Creativity I am someone who... ...makes connections & distinctions between ideas & things 0.773 ...is able to grasp abstract ideas & focus my attention on those ideas 0.805 ...is able to put old information, theories, & so forth together in a new way 0.801 ...uses the materials around me & makes something unique out of them 0.704 Variance 2.381 Proportion 0.595 Cumulative 0.595 Cronbach's Alpha 0.771 85 TABLE 2 Personality Traits: Survey Questions, Factor Loadings, Variances and Cronbach’s Alphas Personality Traits Factor Analysis Questions Agreeableness Conscientiousness Openness to Experience I sympathize with others' feelings 0.856 I am not interested in other people's problems. (reverse coded) 0.715 I feel others' emotions. 0.784 I am not really interested in others. (reverse coded) 0.738 I get chores done right away. 0.615 I often forget to put things back in their proper place. (reverse coded) 0.786 I like order. 0.555 I make a mess of things. (reverse coded) 0.753 I am not interested in abstract ideas. (reverse coded) 0.839 I have difficulty understanding abstract ideas. (reverse coded) 0.784 I do not have a good imagination. (reverse coded) 0.640 Variance 2.440 1.883 1.852 Proportion 0.222 0.171 0.168 Cumulative 0.222 0.393 0.561 Cronbach's Alpha 0.785 0.620 0.647 86 TABLE 3 Descriptive Statistics and Correlation Coefficients (N=678) Variables Mean S.D. 1 2 3 4 5 6 7 8 9 10 1 Opportunity Exposure 0.776 0.417 2 Opportunity Recognition 0.528 0.500 0.1506 3 Opportunity Implementation 0.319 0.466 0.0943 0.6465 4 Firm Foundation 0.438 0.497 -0.0030 0.2333 0.2640 5 Intrinsic Motivation -2.39e-09 1 0.3163 0.1137 0.0128 -0.0316 6 Creativity 3.69e-09 1 0.2459 0.2061 0.1426 0.1805 0.3530 7 Extrinsic Motivation -2.13e-10 1 0.0551 0.0208 0.1194 0.1036 -0.0000 0.1322 8 Openness to experience 4.14e-09 1 0.2103 0.0833 0.0769 0.1462 0.2477 0.5034 -0.0430 9 Agreeableness -2.35e-09 1 0.1630 0.0065 -0.0635 -0.0161 0.1737 0.2078 0.0737 -0.0000 10 Conscientiousness 8.98e-10 1 -0.0370 0.0505 0.0610 0.0095 -0.0622 0.0261 0.0405 0.0000 -0.0000 11 Occupational Relatedness 2.994 1.424 0.1370 0.1643 0.1987 0.1144 0.2374 0.1095 0.1093 0.0705 0.0448 0.0276 12 Contribution 0.906 0.293 0.1893 0.1697 0.0692 0.0410 0.2455 0.1072 0.0028 0.1140 0.0736 -0.0338 13 Mean centered age -0.398 9.327 -0.0153 0.1108 0.1114 0.2395 -0.0062 0.1507 0.0178 0.1350 -0.0928 0.0027 14 (Mean centered age)2 87.023 164.412 -0.0196 0.0570 0.0454 0.0785 0.0386 0.1157 0.0490 0.0496 -0.0543 -0.0123 15 Occupation enjoyment 0.621 0.486 0.0976 0.0774 0.1232 0.0893 0.1973 0.0954 0.0202 0.0286 0.1066 0.0489 16 Female 0.265 0.442 -0.1253 -0.1207 0.0047 -0.0326 -0.1536 -0.2251 0.0156 -0.2151 -0.0205 -0.0384 17 Married/relationship 0.419 0.494 0.0908 0.0841 0.0483 0.1120 0.1156 0.1844 -0.0520 0.1400 0.0333 0.0018 18 Children 0.171 0.377 0.0470 0.0686 0.0760 0.1357 0.0832 0.1574 0.0127 0.1230 -0.0527 -0.0518 19 Region 0.562 0.497 0.0885 0.1300 0.0997 0.1983 0.1771 0.3097 0.1706 0.1759 0.0184 0.0163 20 IT Industry 0.739 0.440 0.1072 -0.0104 -0.0332 -0.1047 0.1744 -0.0822 -0.0249 0.0207 -0.0708 -0.0199 11 12 13 14 15 16 17 18 19 20 11 Occupational Relatedness 12 Contribution 0.0767 13 Mean centered age -0.0695 -0.0610 14 (Mean centered age)2 -0.0322 -0.0213 0.6038 15 Occupation enjoyment 0.5674 0.0389 -0.0052 0.0156 16 Female -0.0938 -0.1600 -0.0716 -0.1477 -0.1223 17 Married/relationship 0.0582 0.0185 0.2797 0.1352 0.1026 -0.2193 18 Children 0.0101 0.0395 0.4944 0.2973 0.0563 -0.1933 0.4319 19 Region -0.0455 0.1725 0.2131 0.0819 0.0148 -0.0953 0.1230 0.1880 20 IT Industry 0.1911 0.1067 -0.0573 -0.0102 0.0893 -0.1598 -0.0671 -0.0153 -0.1728 87 TABLE 4 Determinants of Opportunity Exposure, Recognition, Implementation and Firm Foundation Model 1 Model 2 Model 3 Model 4 Model 5 Variables Opportunity Exposure Opportunity Recognition Opportunity Implementation Firm Foundation Firm Foundation Intrinsic Motivation 0.439*** 0.013 -0.184+ -0.319** (4.04) (0.13) (-1.78) (-3.14) Creativity 0.208+ 0.395*** 0.307** 0.229* (1.65) (3.71) (2.61) (1.98) Extrinsic Motivation 0.120 -0.064 0.197* 0.153+ (1.07) (-0.76) (2.07) (1.67) Openness to experience 0.264* -0.157 -0.004 0.191* 0.166 (2.21) (-1.63) (-0.03) (2.16) (1.57) Agreeableness 0.273** -0.090 -0.223* -0.044 -0.050 (2.61) (-1.05) (-2.43) (-0.53) (-0.58) Conscientiousness -0.080 0.101 0.108 -0.001 -0.036 (-0.75) (1.24) (1.25) (-0.01) (-0.42) Occupational Relatedness 0.089 0.283*** 0.326*** 0.226** 0.232** (0.97) (3.96) (4.18) (3.06) (3.00) Contribution 0.688* 1.171*** 0.713* 0.239 0.415 (2.14) (3.76) (2.20) (0.83) (1.41) Mean centered age -0.002 0.0324* 0.031* 0.067*** 0.066*** (-0.16) (2.51) (2.18) (4.74) (4.50) (Mean centered age)2 -0.001 -0.000 -0.001 -0.001* -0.002* (-0.88) (-0.67) (-1.01) (-2.13) (-2.16) Occupation enjoyment -0.043 -0.204 0.168 0.101 0.176 (-0.17) (-1.01) (0.74) (0.49) (0.83) Female -0.118 -0.306 0.314 0.047 0.099 (-0.51) (-1.54) (1.48) (0.23) (0.48) Married/relationship 0.190 0.070 -0.021 0.058 0.104 (0.76) (0.37) (-0.10) (0.31) (0.54) Children 0.004 -0.231 0.067 -0.098 -0.092 (0.01) (-0.83) (0.23) (-0.36) (-0.34) Region -0.031 0.191 0.142 0.542** 0.503** (-0.13) (1.04) (0.70) (3.05) (2.71) IT Industry 0.391+ -0.191 -0.226 -0.532** -0.392+ (1.66) (-0.95) (-1.05) (-2.67) (-1.96) Constant 0.315 -1.491*** -2.510*** -1.027** -1.370*** (0.72) (-3.76) (-5.89) (-2.76) (-3.50) Number of observations 678 678 678 678 678 Log-likelihood -308.6647 -430.0815 -388.4086 -420.8567 -412.4210 Wald chi2 98.94 68.24 71.78 71.91 76.05 Pseudo R2 0.1445 0.0828 0.0846 0.0944 0.1126 + p<0.1, * p<0.05, ** p<0.01, *** p<0.001; t-statistics in parentheses 88 Chapter 3 MAVERICKS, MISSIONARIES, MASTERMINDS: FOUNDER IDENTITY AND ENTREPRENEURIAL EXIT by Maria Halbinger ABSTRACT Previous research suggests that founders associate meanings with entrepreneurial activities and pursue firm strategies which are consistent with their identities. This study links the characteristics underlying founder identity with key strategic decisions in entrepreneurship and proposes that exit rates are based on founders’ concept of self. The study draws on a unique dataset of 678 individuals entrepreneurially active in hacker- and makerspaces. The findings suggest that founders who are more curious, community oriented and active problem-solvers (i.e. mastermind identity) experience lower exit rates, while founders who are more eager to contribute to the world (i.e. missionary identity) experience higher exit rates. Founders who are more reckless and rivalrous (i.e. maverick identity) are less likely to exit only when firm foundations are recent. By applying a social identity perspective, this study aims to contribute to an understanding of how exit rates of firms vary based on different founder identities. 89 INTRODUCTION Firm exit is a fundamental phenomenon of entrepreneurship that is particularly widespread among young firms (e.g. Brush et al., 2008; Wennberg et al. 2010; DeTienne & Cardon, 2012). Around 60 percent of new firms exit within the first five years (Kirchhoff, 1994; Levie et al., 2011). Since the substantial economic benefits related to economic growth (Sternberg & Wennekers, 2005) and innovation (Schumpeter, 1934) are only realized if new organizations survive (Bruederl et al., 1992; Yang & Aldrich, 2012), it is essential to understand what influences firm exit. This topic has received substantial attention from various research fields. While economics related research focuses largely on the role of the environmental factors on firm exit (e.g. Audretsch, 1995; Klepper, 1996; Geroski et al., 2010), organization scholars engage in a valuable debate over whether survival is determined by market selection mechanisms (Hannan & Freeman, 1989) or adaptation processes (Meyer & Rowan, 1977). Micro level studies emphasize the importance of founders’ human capital, such as prior work experience and demographic characteristics, as crucial aspects of firm performance including firm survival (e.g. Bruederl et al., 1992; Cooper et al., 1994). However, despite their important contributions, these research streams do not take into account that entrepreneurship is not just an occupation. Founders may associate values with their firms and pursue entrepreneurial activities for a personal reason and as a way to convey their identity (Meyer & Zucker, 1989). As a consequence, several studies apply an identity perspective to explain the transition to entrepreneurship (e.g. Dobrev & Barnett, 2005) and provide important insights into how identities relate to entrepreneurial activity thereby shaping the structure of markets and society (Carroll & Swaminathan, 2000; Greve et al., 2006; Sine & Lee, 2009). However, most research in this vein 90 takes a sociological perspective and focuses more on the roles of individuals related to occupation and entrepreneurship (e.g. Ibarra, 1999; Cardon et al., 2009; Sheperd & Haynie, 2009; Hoang & Gimeno, 2010) which puts less emphasis on the social aspect of identity stemming from the field of psychology. Moreover, despite the substantial theoretical investigation, the empirical evidence around the impact of social identity is limited. This chapter adopts the perspective of social identity, i.e. how individuals identify themselves with reference to others or social categories (Tajfel & Turner, 1979; Turner, 1985; Gioia, 1998). This is a useful lens for the investigation of firm exit because founders shape their firms with their identities, and develop firm strategies that are in line with their concept of self (Fauchart & Gruber, 2011). In turn, firm initial strategies have an imprinting effect on firm success (Kimberly, 1979) and firm survival (Delmar & Shane, 2004). Adopting a social identity perspective is particularly advantageous because it allows taking into account founders’ behaviors towards others, which is a crucial aspect since entrepreneurship involves social interaction, and firms are socially constructed (Whetten & Mackey, 2002). Specifically, the present study examines empirically how different founder identities influence firm exit, defined as business termination (Yang & Aldrich, 2012). Different identity dimensions are thus at the heart of this study. I introduce three notions of identity; maverick, missionary and mastermind. Mavericks are founders who are more reckless and inclined to rivalry. Missionaries are founders who are more eager to make a contribution to society and masterminds are founders who are more curious, community oriented and active in problemsolving. It is important to note that identities are not mutually exclusive and individuals may identify equally with each category (Ashforth and Mael 1989). I hypothesize that these three founder identities have different impacts on firm exit, and develop the underlying mechanisms 91 based on the influence that identity has on three core strategic entrepreneurial decisions, i.e. customer needs addressed, targeted markets and resources allocated (Abell, 1980). The present study draws on survey data from 678 individuals in hacker and maker communities in the United States, Canada, Australia, New Zealand, and Northern and English and German speaking Europe. The data were collected through an online questionnaire administered between May and July 2012, and provide individual- and firm-related information, allowing me to distinguish between effects operating at the two levels of aggregation. This context is advantageous because since the sample is based on individuals who are active at different points in the entrepreneurial process, I can analyze exit rates conditioned on the transition to entrepreneurship. Moreover, the setting of a distinct community reduces unobserved heterogeneity and potential bias allowing me to disentangle the effects of different founder identities. By taking advantage of a longitudinal dataset, I find that exit rates are higher for founders who feel more obligated to contribute to the world (i.e. missionaries) and lower for more curious, community-oriented founders (i.e. masterminds). Additionally, when focusing on recent firm foundations only founders who are more reckless and inclined to rivalry, i.e. mavericks, exhibit lower exit rates. These findings contribute to the emerging research that adopts a theoretical perspective on social identity to entrepreneurial phenomena. The findings suggest that founders who identify strongly with their group, i.e. masterminds, can achieve mutual benefits that strengthen the chances of survival of their firms. This supports the notion that users are interesting sources of innovations that can result in firm foundation (Shah and Tripsas 2007; Autio et al. 2013). It shows also that the prototypical individual who subscribes to the cooperative norms, values, and beliefs underlying the maker movement, co-creates products and services of high market potential 92 (Von Hippel 1986; Lilien et al. 2002) and can mobilize important resources relevant for firm survival. The following section introduces the notion of firm exit and founder identity. Section three presents the hypotheses. Section four informs about data and method and Section five presents the results. The chapter concludes with a discussion of the findings and the implications for future research. FIRM EXIT AND FOUNDER IDENTITY The analysis of firm exit has been approached from different perspectives, and scholars have used various terms including survival, failure rate or death. This study applies the notion of firm exit as a measure of performance in that it captures the discontinuation of the firm, and the termination of the business (Yang & Aldrich, 2012). Firm exit is particularly common within the first three years of a venture’s life (Bruederl et al., 1992). However, research on the event of exit events is not fully explored (DeTienne, 2010; Wennberg et al., 2010; DeTienne & Cardon, 2012). Consequently, it is important to investigate the determinants of firm exit, in particular because the benefits such as innovation and technological change (Schumpeter, 1934; Audretsch, 1995), regional development (Feldman, 2001), and economic growth (Sternberg & Wennekers, 2005; van Stel et al., 2005) can only be realized if the firm persists (Bruederl et al., 1992; Yang & Aldrich, 2012). Firm exit has been investigated in various studies related to economics and organization research. The economics literature associates firm exit rates with industry specifics including industry lifecycle and innovation rates (e.g. Audretsch, 1995; Klepper, 1996) as well as human capital related factors (Bruederl et al., 1992; Cooper et al., 1994; Dahl & Reichstein, 2007). 93 Organization research investigates firm exit from two different perspectives. One research stream highlights the role of learning and adaptation as important factors in survival (e.g. Meyer & Rowan, 1977); the other, which is related to organizational ecology, studies the birth and mortality rate of organizations and attributes exit or failure rates to market selection mechanisms rather than (the lack of) adaptation by the focal organization (Hannan & Freeman, 1989; Carroll & Hannan, 2000). These research strands have significantly increased our understanding of the dynamics of exit at firm level but focus less on the role of the individual entrepreneur, particularly with respect to the meanings and values that founders associate with their firms. Given that most organizations in any industry sector are rather small (Carroll & Hannan, 2000), researchers have emphasized the need to take account of the role of the individual founder when analyzing firm exit (Bruederl et al. 1992). The founder may represent a critical source of variation (Wennberg, 2009) and ignoring factors related to firm founders could result in inaccurate inferences (Pennings et al., 1998). Hence, analyzing the role of founder identity could provide valuable insights into the event of firm exit since scholars emphasize the close link between the founder and the firm, considering the firm as the embodiment of the founder’s identity (Meyer & Zucker, 1989). Specifically, founders systematically apply firm strategies that are aligned with their selfconcept and imprint their firms with their identities (Fauchart & Gruber, 2011). At the same time, firm strategy is strongly related to firm outcome including survival (Kimberly, 1979; Delmar & Shane, 2004). Hence, given that an individual behaves in a way that is consistent with his or her identity (Stets & Burke, 2000), one could imagine that, through imprinting, founder identity would also influence firm exit. This suggests that research on exit would benefit from closer attention to founders’ identity. 100 Finally, for missionaries’ self-evaluation, acting socially responsibly is crucial. Their intention is to make the world a better place, and hence, breaking with rules and norms is against their maxims. Missionaries’ moral standards require them to act accordingly even if this limits the scope of their entrepreneurial action. This is because rules cannot comprise all contingencies, and particularly in dynamic, competitive environments such as entrepreneurship, acting outside of the accepted norms, and breaking with the rules are inevitably linked to creativity and are required to generate change and innovation (Brenkert, 2009). However, creativity is strongly associated with opportunity discovery (Baron, 2006) and entrepreneurial action (Amabile, 1997; Ward, 2004), particularly as firm founders typically face unforeseen situations and challenges. Based on these three rationales I posit that: Hypothesis 2: Founders with a stronger missionary identity exhibit higher exit rates. Masterminds Founders inclined towards a mastermind identity show lower exit rates for three reasons. First, masterminds’ frame of reference is a community to which they feel a belonging. Consequently, market offerings that are created in mastermind firms are targeted at satisfying the needs of the community. Since beliefs and values are shared within a community (Tajfel & Turner, 1979), masterminds are aware of members’ requirements and likely have similar demands, and similar problems with existing market offerings like products or services. As a result, when masterminds develop their business ideas, they respond to the community members’ needs and problems. Users are an important source of innovation for firms, particularly in relation to the identification of sophisticated needs ahead of a trend, and the development of potential solutions to address those needs (e.g. von Hippel, 1986). Their solutions and entrepreneurial opportunities are 101 new compared to existing products and services, and exhibit market potential (Lilien et al., 2002). Both aspects can contribute crucially to the firm’s viability since they represent important features of sustainable, competitive advantage (Barney, 1986; 1991). The developments generated by individuals associated with communities tend to be innovative, a phenomenon that is evident across several industries and professions including librarians (Morrison et al., 2004), sports equipment industry (e.g. Franke & Shah, 2003), and open software development (e.g. von Hippel & von Krogh, 2003; Alexy & Leitner, 2011). Prior work has shown that their entrepreneurial activities resemble a trial and error process (e.g. Lee & Cole, 2003) that they have potential for firm creation (Shah & Tripsas, 2007). Knowing about user needs and potential technological solutions introduces business-relevant effectiveness into the entrepreneurial process. Masterminds are able at an early stage to develop market offerings that are clearly targeted to a particular group and which can be beneficial to stay in the market. Second, as masterminds strongly identify with the group and its members’ values, supporting and being supported by the community is core to their social motivation. Since they are able to support their fellows in their entrepreneurial activities, masterminds receive reciprocal help from the community. Thus, peers represent a highly valuable network, and the resources of the community are an important source of innovation and prototyping activities (e.g. Franke & Shah, 2003; Lee & Cole, 2003). Since community members can be mobilized to support the development of new market offerings and serve as pilot customers to test products or service features before market launch, masterminds reap substantial benefits for entrepreneurship which increase their efficiency and the firm’s viability. Third, mastermind-prone founders’ self-evaluation depends on the extent to which they are able to contribute to their community. Being an asset to the group matters, particularly as being 102 valued is important for these individuals’ self-worth and feeling of belonging (Hogg & Terry, 2000). Hence, masterminds aim consistently to provide their peers with beneficial market offerings in order to win their appreciation; hence, maintaining their business is important for their identity. To achieve this, masterminds thrive to improve their skills and capabilities through learning. In turn, learning, for instance in the form of probing, provides masterminds with important information advantages in relation to technological developments and their viability (Autio et al., 2013). In other words, masterminds learn earlier about the existence and quality of a business opportunity which decreases their exit rates. Hypothesis 3: Founders with a strong mastermind identity exhibit lower exit rates. DATA AND METHOD This section describes the empirical setting and the dataset used to test the hypotheses, and introduces dependent and independent variables and methods. Empirical Setting Survey data were collected through an online questionnaire administered between May and July 2012 to members of so-called hacker- and makerspaces. Hacker- and makerspaces are communities of individuals who share an interest in technology, computing, science, art and the hacking culture in general, and where free flow of information, transparency, and learning represent core values (Coleman & Golub, 2008). An initial subscription and voluntary fee payments allows access to these communities which typically offer physical workspaces equipped with machinery such as 3D printers, laser cutters, computer numerically controlled machines, and tools such as hammers and saws, for shared use. While these workspaces can be physically located in warehouses, basements, or community centers and facilitate cooperation, socializing, sharing of 103 ideas and resources, most hackerspaces are represented by virtual collaboration. On joining a hacker- and makerspace, an individual is included in an email distribution list which allows exchange of various information on needs or problems, solutions and ideas related to hacking projects. Hackers and makers are individuals who engage in development and problem-solving activities, thereby not only modifying the work of others but also creating new solutions, products, services or artifacts (Lang, 2013; Aldrich et al., 2014). These activities are pursued individually, within and across hackerspaces, and typically motivated by curiosity and personal interest, and range from writing code and programming to the development of products, drones, robots, and public performances in association with light installations, music, or fire. Hackers typically are well-educated or are pursuing a degree or higher degree course (Lakhani & Wolf, 2005), and rather than working in organizational hierarchies prefer occupations that allow personal freedom including entrepreneurship (Carlson, 2011). The context was chosen to investigate the research question because it represents a setting where all the required variables coexist thereby constituting three major benefits. First, hackers and makers engage in diverse kinds of entrepreneurial activities, including but not limited to opportunity discovery and exploitation. This is especially beneficial for an analysis of exit rates, because not all discovered opportunities result in the creation of firms; thus, it is possible to study firm exit conditioned on the transition from opportunity discovery to entrepreneurship. Second, hacker- and makerspaces are beneficial for investigating identity because their members represent a reasonably homogeneous population that can be observed within a community with distinct boundaries. This allows the collection of data on fairly comparable individuals, whose main values and ideologies are shared and the community norms are institutionalized. A more 104 diverse community would likely introduce a higher degree of heterogeneity across participants and potentially produce more biased results. This setting offers major advantages to this respect and responds to calls in entrepreneurship research for settings that include entrepreneurs and nonentrepreneurs who are fairly comparable individuals (Shane & Khurana, 2003). Third, although hackers and makers share some basic features, there is substantial heterogeneity with respect to founder identities. Since the setting represents a particular community of hackers and makers, it is possible to single out differences across identities within the overall community. This allows observation of key aspects of identity that lead to individual differences on different identity dimensions, such as the social motivation underlying their entrepreneurial activity. Data Collection and Final Sample An iterative process, and a combination of field observations, interviews and test studies was used to achieve ex ante insights into the setting (Brewer, 2000; Barley & Kunda, 2001) and facilitated derivation of an appropriate research design. The survey was designed based on single- and multiple-choice questions, and divided into firm-level and individual-level information including demographics, motivations for hacking, and entrepreneurial engagement. The questionnaire was tested in several offline and online pilots. The data were collected in three steps. First, the administrator of a worldwide hacker and maker consortium 8 marketed the survey on three social media platforms to gain reliability within the hacker community because the field observations showed that trustworthiness would be a crucial condition for participation in the survey. In the second phase, emails were sent to hacker and makerspaces in the United States, Canada, Australia, New Zealand, and Northern and English and 8 This consortium is the biggest conglomerate of hacker and makerspaces worldwide and was thus, selected for this study 105 German speaking Europe. The spaces were selected based on a) their registered status (active or non-active), b) their accessibility, c) their website claims about their purpose and intentions, d) their conditions of membership. Emails were sent to 392 communities and community organizers asking for the survey to be distributed across the spaces. Among these communities, 5.86% could not be reached, leaving 369 hacker and maker spaces; reminders were sent after 10 days. Imposition of a response time window of 7.5 to 90 minutes on the 2948 clicks of individuals registered on the survey platforms, produced a final sample of 678 respondents. This time window was chosen based on response time measured in the pilot surveys. Respondents not meeting these requirements were assumed to have visited the platform out of curiosity and were excluded from the analyses. 9 For the investigation of entrepreneurial identities, the final data set includes individual- and firm-level information including firms’ market entry and exit dates. In the final sample, 76.6 percent of individuals work on developments related to information technologies, and 43.8 percent of individuals have firm founding experience. In terms of demographics, the sample is characterized by a mean age of 32.9 years, and an overrepresentation of males (498 male versus 57 female respondents) and singles with a 39.52 percent rate. To check the representativeness of the final sample, various tests were conducted with respect to the demographic dimensions of age and gender as well as the distribution of respondents across hackerspaces and regions. The final sample appears to correspond to the characteristics of the overall community and with research in similar settings (e.g. Lakhani & Wolf, 2005; Jeppesen & Lakhani, 2010). Although, the distribution of the entrepreneurship rate across regions is in line 9 This choice was investigated by relaxing the minimum time threshold to 8.5 minutes and 10 minutes but keeping 90 minutes as the maximum time. An individual taking more than 90 minutes would seem to indicate gaps in responding which could bias results. With both alternative specifications, the main results hold. 106 with the Global Entrepreneurship Monitor (GEM) 2013 report (Amoros & Bosma, 2014), there is indication that hackers and makers have a higher proclivity for entrepreneurship than the general population, which suggests that the findings may not reflect a representative individual. The results did not change for the checks for potential bias based on hackerspace affiliation or region. Measures Dependent variable The dependent variable exit, is a dummy variable that takes the value 0 if the firm did not exit the market within 3 years and 1 otherwise. In this study, exit relates to closure of the business (Yang & Aldrich, 2012). The exit variable was generated based on the questions “When did you (co-) found your most recent company?” and “When did you close down your most recent company?”. Consequently, there is no strong reason to believe that the variable would capture an exit route other than business termination, for instance exit through buyout. Explanatory variables To generate the explanatory variables, I follow prior work in entrepreneurship research and use previous studies as conceptual guidelines to create and mark the domain of these variables (e.g. Brown et al., 2001). Founder identity is based on three sub-dimensions (Brewer & Gardner, 1996) which provided guidelines to generate the scales of the founder identity variables. I conducted principal component factor analyses to operationalize the variables based on the underlying threedimensional structure of: a) basic social motivation, b) self-evaluation, and c) frame of reference. Hence, each factor analysis was based on three survey items, one item for each aspect of identity. In this way, overall nine questions, which were measured on a Likert scale from 1 (“Strongly disagree”) to 7 (“Strongly agree”), were used for the factor analyses supplemented by the application of the Kaiser (1960) criterion and varimax rotation. 107 Each factor represents a founder identity variable, 1) maverick identity, for rather reckless founders with an affinity for competitive behavior, 2) missionary identity, for founders engaging in entrepreneurship primarily in order to make a contribution to society, and 3) mastermind identity, for more curious, community-oriented founders active in problem-solving. Importantly, these are not mutually exclusive categories. In this respect, the operationalization of identities departs from Fauchart and Gruber’s original founder identity typology (2011) and refers specifically to their additional finding that beyond pure types, identities can blend in the sense that so-called hybrid identities can occur. To explore this remarkable finding, allowing more degrees of freedom is crucial and therefore, in the present study scales representing a continuum along which individuals can vary, are applied. Using factor analysis to create the variables is particularly beneficial because it increases our understanding of the cross-correlations with multiple variables in the data set. An overview of the items used per identity dimension is presented in Table 1, descriptive statistics and factor loadings are presented in Table 2. ------------------------------------------ Insert Tables 1 and 2 about here ------------------------------------------- Control variables The control variables are basically categorized as individual or firm and industry-related. Individual level controls Identity is associated with actions, beliefs, and motivation (Hogg & Terry, 2000). Thus, individuals’ particular goals and the motivations that drive their entrepreneurial actions might be relevant for an analysis of different types of identity. The dummy control variable goal orientation is generated based on the question “I am someone who is motivated by goals” taking the value 1 for 108 “Strongly agree”, “Agree” “Weakly agree” and 0 for “Strongly disagree”, “Disagree” and “Weakly disagree” and “Neither nor”. The controls for extrinsic motivation and intrinsic motivation were created based on factor analysis capturing whether the individual engages in entrepreneurial-related activities based on an inherent interest in the activity itself, or based on external, outcome driven expectations. Internal consistency measured by Cronbach’s alpha is high for extrinsic motivation (alpha= 0.729) and poor for intrinsic motivation (alpha= 0.478). With regard to the particular context, individuals’ participation and interest in gaining reputation matter (e.g. Roberts et al., 2006). Thus, the dummy variable, contribution, measures whether or not the individual actively participated in the community, and reputation is operationalized based on the question “I hack because I want to enhance my reputation/status in the community” (taking the value of 1 for “Weakly agree”, “Agree”, “Strongly agree” and 0 for “Strongly disagree”, “Disagree”, Weakly disagree”, “Neither nor”). Furthermore, innate personality traits can play a role in individuals’ identity with respect to the groups to which they feel a belonging (Weber et al., 2011; Sibley & Duckitt, 2008). Thus, the analysis also includes dimensions of the five-factor model of personality (Costa & McCrae, 1992). This study applies a short scale to measure dispositions (Donnellan et al., 2006), which is in line with prior work in management, (e.g. Grant & Berry, 2012). Since conscientiousness, agreeableness, and openness to experience are relevant in entrepreneurship studies (e.g. Zhao & Seibert, 2006), these three dimensions are included in the analysis. Principal component factor analysis allows exploration of the questions’ factor loadings into the distinct dispositions, supplemented by application of the Kaiser (1960) criterion and varimax rotation. The Cronbach’s alphas for the personality traits - agreeableness (alpha= 0.785), conscientiousness (alpha= 0.620) 109 and openness to experience (alpha=0.647) - suggest acceptable to good levels of internal consistency. With respect to the ongoing debate over whether disposition and identity are stable over time, the analysis includes the control growth mind. This variable captures whether individuals have a fixed mindset reflecting their belief in stable, innate abilities (taking the value 0) or a growth mindset represented by their belief that they can change in the sense that abilities and characteristics can be developed (taking the value 1) (Dweck 1999; 2006; Grant & Dweck, 2003). To construct this variable, survey items from social psychology were adapted. The variable is based on factor analysis of the following three, relative measurements “If I knew I wasn’t going to do well at a task, I probably would not do it even if I might learn a lot from it.”, “I sometimes would rather do well than learn a lot” and “It is much more important for me to be challenged than it is to demonstrate my intellectual ability” (alpha=0.50). Since creativity can influence occupational performance, particularly in entrepreneurial settings (Amabile, 1997; Amabile et al., 2005), the factor is included as a control variable, and generated based on factor analysis with Cronbach’s alpha of 0.77 indicating a good level of internal consistency. Age affects entrepreneurship, both in its linear and curvilinear forms (e.g. Dunn & Holtz- Eakin, 2000, Oezcan & Reichstein, 2009) and is included in the present analysis as mean centered age and [mean centered age]2. In line with prior work, demographic dummy variables such as gender (with 1 indicating female), being in a relationship (married/relationship) and children are also included (e.g. Dunn & Holtz-Eakin, 2000; Sørensen 2007). Moreover, the binary variable occupation enjoyment which 116 addition the firm and industry related control variables. Model 4 introduces the explanatory variables maverick, missionary and mastermind identity and contains all the effects that are hypothesized in this chapter. The Heckman specification in Model 1, performed by probit regression, shows that opportunity implementation is highly associated with the transition to entrepreneurship. This suggests that individuals who have already implemented an opportunity (“hacks”) in the form for instance a product or service, are more likely to become entrepreneurs. Hypothesis 1 stating that individuals with a stronger maverick identity exhibit lower exit rates is not confirmed. As shown in Model 4, maverick identity and entrepreneurial exit are negatively associated but not significantly. Hypothesis 2 states that founders with a stronger missionary identity demonstrate higher hazards of entrepreneurial exit. In support of the hypothesis, Model 4 presents the positive significant effect of the missionary variable on entrepreneurial exit indicating that ventures founded by missionaries have higher propensities to exit the market. The analysis also supports Hypothesis 3, that a stronger mastermind identity is negatively associated with the hazard of entrepreneurial exit. The mastermind identity variable exhibits significantly negative association with entrepreneurial exit. The significant negative mastermind variable in the Heckman regression in Model 1 also suggests that individuals’ propensity for mastermind identity inhibits transition to entrepreneurship. This suggests that individuals with high scores on mastermind identity are less likely to become entrepreneurs but if they do so, they are more likely to survive (the analysis shows lower hazards of exit). These findings highlight the importance of modeling entrepreneurial exit conditioned on the 117 transition to entrepreneurship since the effect of the mastermind variable might otherwise suffer from bias. Model 1 also shows that creative individuals are more likely to become entrepreneurs; individuals with a growth mindset seem to be predisposed to transition to entrepreneurship. Being an entrepreneur and the enjoyment of occupation also show a positive association. While being extrinsically motivated has a positive effect on the transition to entrepreneurship, being intrinsically motivated seems to inhibit the transition. Model 4 highlights other determinants of entrepreneurial exit. While both location in the Anglo-Saxon region and firm foundation during the years of disruption increase the hazard of entrepreneurial exit, being a parent seems to have the opposite effect. Supplementary Analyses Several supplementary analyses were conducted to test the robustness of the results (see Table 6). First, I ran the analysis on a sub-sample of recent entrepreneurs with firm foundations in 2005 and later. After applying the same procedure of the two-stage Heckman specification to measure firm exit at the three year threshold, Model 5 shows that the results are even stronger than expected and that I underestimated the effects of identity in the main analysis. All the hypotheses are supported since both masterminds and mavericks have a negative and highly significant association with firm exit (p<0.01) while the missionary variable is positively linked to firm exit (p<0.05). Additionally, Model 5 shows that the variables creativity and intrinsic motivation are positively and significantly linked to firm exit. In contrast, the personality trait agreeableness and having children are negatively related to firm exit. 118 Second, to rule out the possibility that the results are a by-product of model choice, I consider time to be continuous, and apply a Cox (1972) proportional hazard model (Jenkins 2005). As shown in Model 6, the results are similar to the findings from the main analysis (Model 4 Table 5). This suggests that the results are robust independent of model selection. The final supplementary analysis aims to take into account of whether the identity effect is rather short than long term by examining firm exit at a five year threshold, as shown in Model 7. Since over a half of new ventures exit the market by the fifth year after foundation, as for example reported by Levie and colleagues (2011), this time span was chosen to test the hypotheses further. The supplementary investigation was conducted following the procedure described in the main analysis including data expansion based on a five year period and the two stages, a Heckman specification and a discrete time duration model. With respect to the five year threshold, all observations for firms founded later than 2007 were disregarded to avoid censoring issues. Model 7 shows that all the explanatory variables point in the direction as hypothesized but only the association with the mastermind variable holds significantly as predicted. One explanation for these findings might be that identity effects are stronger in the upper tail of the hazard ratio distribution (higher rates of exit) and that identity matters less on lower rates of exit. In other words, those study subjects that are left in the distribution exhibit less variation, and consequently, identity becomes less important. More detailed explanation of these findings is provided below. Further results in Model 7 show that openness to experience, reputation, gender, being in a relationship, disruption, and Anglo-Saxon region, increase the likelihood of exit, while having children decreases the likelihood of exit. Finally, all models show significant results for age, either mean-centered or the squared term. The significance of the inverse mills ratio variable across all models suggests that the 119 Heckman specification is the appropriate methodological procedure for all three supplementary analyses. 10 ------------------------------------------ Insert Tables 5 and 6 about here ------------------------------------------- DISCUSSION The findings of this study indicate that founder identities affect firm exit in different ways. Specifically, founders that are more community oriented, active in problem-solving and learning (i.e. mastermind identity) are negatively associated with firm exit while the opposite effect is found for founders that are more eager to contribute to the world (i.e. missionary identity). In particular, the supplementary analysis on a sub-sample of recent firm foundations supports all the hypotheses suggesting that both maverick and mastermind identities are highly significantly and negatively associated with firm exit, and that missionary identities show the opposite effects. The results support research on firm exit by investigating business termination from a founder identity perspective as grounded in social psychology. This research line suggests that firms are an expression of the founders’ identities, and since values, beliefs, and actions are consistent and confirm identity, the firm’s strategic actions are inevitably linked to the founder (Hogg & Terry, 2000; Fauchart & Gruber, 2011). The present study extends this view by theoretically intertwining aspects of identity and core decisions in entrepreneurship to theorize how 10 In a fourth, altered model I included a control for revenues accumulated in the first 2 years after foundation to analyze firm exit based on the three year window. The effects are significant and as hypothesized for mavericks and masterminds indicating that these founder identities are negatively associated with firm exit independent of firm revenue. The missionary identity variable shows a negative significant association with exit and is hence in full contrast to the hypothesis. This indicates that if revenues are held constant, also missionaries are less likely to exit. However, due to data restrictions, this analysis is based on only 29 observations which can be problematic in terms of the stability of the findings. 120 exit rates, like firm strategy, vary based on identity. Moreover, by conceptually and methodologically applying the notion of founder identity as a dimension along which individuals can vary, the study acknowledges that in most industry settings, hybrids rather than pure identity types prevail and are likely to increase in the future (Fauchart & Gruber, 2011). In addition, the supplementary analyses contribute to the ongoing debate on whether identity is stable over time. While the maverick variable loses significance in the full sample, all the hypotheses are strongly supported by using a sub-sample of recent firm foundations. The consistent findings for missionaries and masterminds confirm previous research by implying that these two founder identities are particularly well developed and that actions are tightly intertwined with identity (Fauchart & Gruber, 2011). Moreover, the negative association between mastermind identity and the transition to entrepreneurship in the first stage of the empirical analysis, complements work on social identity and user entrepreneurship. Research on user entrepreneurship suggests that users become entrepreneurs by accident, in the sense that venture creation is often not the intention when they alter existing market offerings or generate new products. Instead it is the exposure of the innovation to the public that may initiate the idea of commercializing the innovation by starting a firm (Shah & Tripsas, 2007). For community-oriented individuals such as masterminds however, it appears to be particularly important to be in line with community values which typically incorporate free information exchange and free usage of the innovation developed, as in communities related to sports equipment (e.g. Shah, 2000) and open software developments (von Hippel & von Krogh, 2003; O’Mahony, 2003). In these communities, public exposure is also used, but on purpose in the sense that innovation-related information is publicly revealed in order to prevent third parties from appropriating the innovation. This corresponds to prior work showing that in order to guarantee 121 benefit from the innovation for everyone outside the typical corporate appropriation regimes, other forms than firm foundation are typically used to protect the works of community members, for instance transfer of knowledge into non-profit foundations (O’Mahony, 2003). Moreover, it is consistent with the notion that large corporations and communities may have contrasting rationales for their existence thereby putting their interactions in tension, and rendering some relationships parasitic (Dahlander & Magnusson, 2005). This would suggest that exploiting a development generated within a community in the form of firm foundation rather contrasts with community values. Hence, firm foundation may be depreciated and even stigmatized, and considered as “joining the other side”. Accordingly, from a social identity point of view, a community member’s transition to entrepreneurship represents atypical behavior and deviance from the group. Along these lines, past research implies that in becoming an entrepreneur, founders lose their feeling of belonging to the group, which is a drawback – the so-called “dark side of entrepreneurship” (Sheperd & Haynie, 2009). However, the finding related to the negative link between a mastermind identity and firm exit once the transition to entrepreneurship has been made, extends this view and is consistent with the “overachiever” argument (Marques, 1990; Wann et al., 1995). In line with this argument, individuals deviate in a positive way as they become “highflyers” in the group (Hogg & Terry, 2000). Since masterminds serve the community with useful market offerings, they turn into highly valued, and therefore appreciated members of the group which feeds their feeling of belonging. Hence, masterminds who become entrepreneurs face a rather “bright side of entrepreneurship” which strengthens their community-oriented identity. Furthermore, the findings of the supplementary analysis to check short versus long term effects of identity suggest several possible explanations. First, the significant negative association 122 of the mastermind variable with firm exit given a five year window, provides strong grounds for the assumption that the mastermind identity is stable within a long term perspective. This suggests that user-innovators may create firms indeed “accidentally” (Shah & Tripsas, 2007) but that these firms have higher survival chances within a long term perspective. Second, the loss of significance for maverick and missionary might indicate that these identities are more inclined to take professional teams on board within the first few years of firm foundation. While masterminds draw on their community to access resources, i.e. development support or management advice, mavericks and missionaries may need to involve external professionals to mobilize relevant human or financial capital, for instance technical experts and managers. At the same time, professionals are likely to act in ways that are consistent with the norms and routines of their professional field (e.g. DiMaggio & Powell, 1983; Colyvas & Powell, 2006). Hence, if a professional team takes over, the firm will pursue actions and strategies in line with the team and it can be assumed that the original founder’s identity and his or her actions will matter less over time. Finally, these findings contribute to the literature on opportunity identification and development (Shane & Venkataraman, 2000; Baron, 2006). As noted earlier, it might be that some identities matter primarily in the upper tail of the distribution when exit rates are high. Once the identities with high likelihoods of exit are ”removed”, there might be a substitution effect between the identity and the quality of the opportunity. In other words, over time, only high-quality opportunities will be left in the market. These opportunities may have been high quality from the start (and not influenced by a less survival-likely identity) or may have been successfully transformed so that their quality has improved over time. These effects seem to apply particularly to 123 for the mastermind identity since the negative association with firm exit is significant and stable in all the analyses. Implications for research Users have been considered important sources of innovation (e.g. von Hippel, 1986; 2005), and examples of innovative developments span industries including sports equipment (e.g. Baldwin et al., 2006), juvenile products (Shah & Tripsas, 2007), and medical devices (Lettl et al., 2008). At the same time, we know that innovation and entrepreneurial action are strengthened through community interaction (Franke & Shah, 2003; Autio et al., 2013). The present study is in line with the notion that hacker- and makerspaces may represent potential infrastructures for users and thereby decrease the barriers to innovation and entrepreneurship (Aldrich et al., 2014). By providing online and physical spaces - often equipped with tools and machinery - hacker- and makerspaces enable opportunity development and facilitate entrepreneurship. Moreover, the study implies that particular identities with close links to these communities can benefit greatly in relation to the various activities involved in the entrepreneurial process including firm survival. The prevailing values of sharing and collaboration are highly beneficial for entrepreneurship because they promote development of business ideas, prototyping, and access to first customers. This phenomenon is interesting from a technological and an institutional perspective, and has interesting implications for research on entrepreneurship and innovation and resource dependency. Implications for practitioners and society This study is relevant to practitioners because it shows that masterminds in particular, develop valuable business opportunities that are not always commercially exploited. Venture 124 capitalists and managers can benefit from this insight as these business ideas can be skimmed from the market to become the foundation for new ventures and to foster inorganic growth and innovation in corporations. Also, awareness of the positive relationship between a missionary identity and firm exit is helpful for practitioners deselecting among ideas for entrepreneurship. Moreover, this study can help to use identity as an instrument to apply a particular strategy. For instance, being aware that founders apply identity distinct strategies, and knowing that for instance a founder has a proclivity for the missionary identity may be helpful alter the strategic directions accordingly. Finally, the study increases our understanding of the different identities and social motivations behind hacking. The term “hacker” suffers from negative connotations with criminal attacks and encroachments. This study, and especially the field work in preparation for the study design, suggests that this notion of the hacker as an individual that illegally breaks into security systems may be one aspect of term. The notion of hacker also includes individuals that legally alter existing products and services and develop new ideas with innovation potential. The present study appears to be the first attempt to empirically analyze these identities and highlight their importance for entrepreneurship, innovation, and society. Limitations and Future Research A natural concern in terms of limitations refers to the generalizability of the study’s findings. The particular context of hackers and makers raises questions about the extent to which the results can be generalized to other groups, entrepreneurs, or individuals. This setting is specific in the sense that collaboration is the baseline, and hence, the results might be less informative in contexts where interaction and collaboration are less important. However, researchers from various 125 disciplines have analyzed individuals’ identities and the impact on firm characteristics (e.g. Hambrick & Mason, 1984; Whetten & Mackey, 2002), as well as the influence on market structure and economic development, based on other specific contexts, such as microbreweries (Carroll & Swaminathan, 2000), windmills (Sine & Lee, 2009), and radio stations (Greve et al., 2006). It should be noted that similarities across fields are evident, and research confirms the influence of missionary identities (e.g. Whetten & Mackey, 2002) or individuals with a more community oriented identity (e.g. Sine & Lee, 2009, Fauchart & Gruber, 2011). The field of entrepreneurship has benefitted crucially from these studies as they have increased our understanding of why and how different identities pursue entrepreneurial activities. Hence, there is no strong reason to believe that the present results would not hold in different settings, in particular in settings where innovation and entrepreneurial activities are in place, for instance think tanks or corporate R&D labs. Previous research has highlighted the importance of using appropriate settings that include non-entrepreneurs comparable to entrepreneurs in order to analyze entrepreneurship on the individual level (Shane & Khurana, 2003). The specific setting of hackers and makers includes entrepreneurs and non-entrepreneurs and thus is beneficial since it allows us to observe and disentangle the effects of different identities in a closed context with comparable individuals to reduce unobserved heterogeneity. Furthermore, previous research on entrepreneurial survival points out that founders have different performance thresholds that can alter exit rates independent of economic reasons (Gimeno et al., 1997). The present study takes this consideration into account and disentangles the identity effects by keeping several survival-related factors constant including industry and firm variables such as firm and team size, as well as time fixed effects. In a supplementary analysis I controlled 132 Jenkins, S. P. 2005. Survival analysis. Unpublished manuscript, Institute for Social and Economic Research, University of Essex, Colchester, UK. Retrieved December 30, 2012, http://www.iser.essex.ac.uk/files/teaching/stephenj/ec968/pdfs/ec968lnotesv6.pdf. Jeppesen, L.B., & Lakhani, K. 2010. Marginality and problem-solving effectiveness in broadcast search. Organization Science, 21: 1016-1033. Jost, J. T. 1995. Negative illusions: Conceptual clarification and psychological evidence concerning false consciousness. Political Psychology, 16: 397-424. Jost, J. T., & Banaji, M. R. 1994. The role of stereotyping in system-justification and the production of false consciousness. British Journal of Social Psychology, 33: l-27. Kaiser, H. F. 1960. The application of electronic computers to factor analysis. Educational and Psychological Measurement, 20: 141–151. Kimberly, J.R. 1979. Issues in the creation of organizations: Initiation, innovation, and institutionalization. Academy of Management Journal, 22: 437–457. Klepper, S. 1996. Entry, exit, growth, and innovation over the product life cycle. The American Economic Review, 86: 562-583. Korosec, R.L., Berman, E.M. 2006. Municipal support for social entrepreneurship. Public Administration Review, 66: 448–462. Lee, G. K., & Cole, R. E. 2003. From a firm-based to a community-based model of knowledge creation: The case of the Linux kernel development. Organization Science, 14: 633-649 Lettl, C., Hienerth, C., & Gemuenden, H.G. 2008. Exploring how lead users develop radical innovation: Opportunity recognition and exploitation in the field of medical equipment technology. IEEE Transactions on Engineering Management, 55: 219-233. Levie J, Gavin D, Leleux B. 2011. The new venture mortality myth. H.K.K. Kim (Ed.), Handbook of research on new venture creation, Edward Elgar Publishing Limited, Cheltenham. Lilien, G. L., Morrison, P. D., Searls, K., Sonnack, M., & von Hippel, E. 2002. Performance assessment of the lead user idea-generation process for new product development. Management Science, 48: 1042-1059. Lounsbury, M., & Glynn, M. A. 2001. Cultural entrepreneurship: Stories, legitimacy, and the acquisition of resources. Strategic Management Journal, 22: 545-564. Marques, J. M. 1990. The black-sheep effect: Out-group homogeneity in social comparison settings. In D. Abrams & M. A. Hogg (Eds.), Social identity theory: Constructive and critical advances: 131-151. London: Harvester Wheatsheaf. 133 McClelland, D. 1961. The achieving society. Princeton, NJ: Van Nostrand. Meyer, J. W., & Rowan, B. 1977. Institutionalized organizations: Formal structure as myth and ceremony. American Journal of Sociology, 83: 340-363. Meyer, M.W., & Zucker, L.G. 1989. Permanently failing organizations. Beverly Hills, CA: Sage. Morrison, P. D., Roberts, J. H., & Midgley, D. F. 2004. The nature of lead users and measurement of leading edge status. Research Policy, 33: 351-362. Nelson, R., & Krashinsky, M. 1973. Two major issues of public policy: Public policy and organization of supply. In R. Nelson & D. Young (Eds.), Public Subsidy for Day Care of Young Children: 47–69, D.C. Heath: Lexington, MA. Oezcan, S., & Reichstein, T. 2009. Transition to entrepreneurship from the public sector: predispositional and contextual effects. Management Science, 55: 604-618. Paton, R. 2003. Managing and Measuring Social Enterprises. SAGE Publications: London, U.K. Pennings, J., M. , Lee, K., & Witteloostuijn, A. v. 1998. Human capital, social capital, and firm dissolution. Academy of Management Journal, 41: 425-440. Powell, W.W. & Sandholtz, K.W. 2012. Amphibious entrepreneurs and the emergence of organizational forms, Strategic Entrepreneurship Journal, 6:94-115. Roberts, J.A., Hann, I.-H., & Slaughter, S.A. 2006. Understanding the Motivations, Participation, and Performance of Open Source Software Developers: A Longitudinal Study of the Apache Projects. Management Science, 52: 984-999. Schumpeter, J.A. 1934. The theory of economic development. Cambridge, MA: Harvard University Press. Schumpeter, J. A., 1975. Capitalism, Socialism and Democracy, 3rd edition. Harper Colophon Books, New York. Shah, S. K., & Tripsas, M. 2007. The accidental entrepreneur: The emergent and collective process of user entrepreneurship. Strategic Entrepreneurship Journal, 1: 123–140. Shane, S.2003. A general theory of entrepreneurship: The individual-opportunity nexus. Northampton, MA: Edward Elgar. Shane, S., & Venkataraman, S. 1996. Renegade and Rational Championing Strategies. Organization Studies, 17: 751-771. 134 Shane, S., & Venkataraman, S. 2000. The promise of entrepreneurship as a field of research. Academy of Management Review, 25: 217-226. Shane, S., & Khurana, R. 2003. Bringing individuals back in: the effects of career experience on new firm founding. Industrial and Corporate Change, 12: 519-543. Shane, S., Locke, E.A., & Collins, C.J. 2003. Entrepreneurial motivation. Human Resource Management Review, 13: 257-279. Shane, S., Nicolaou, N., Cherkas, L., & Spector, T. D. 2010. Genetics, the Big Five, and the tendency to be self-employed. Journal of Applied Psychology, 95:1154-1162. Shepherd, D.A., & De Tienne, D. R. 2005. Prior Knowledge, potential financial reward, and opportuntiy identification. Entrepreneurship Theory and Practice, 29: 91-112. Shepherd, D. A., & Haynie, J. M. 2009. Birds of feather don’t always flock together: Identity management in entrepreneurship. Journal of Business Venturing, 24: 316 –337. Sibley, C. G. & Duckitt, J.. 2008. Personality and prejudice: A meta-analysis and theoretical review. Personality and Social Psychology Review, 12: 248–79. Sine, W.D., Lee, B.H. 2009. Tilting at Windmills? The Environmental Movement and the Emergence of the U.S. Wind Energy Sector. Administrative Science Quarterly, 54: 123– 155. Sternberg, R.J. 1985. Implicit theories of intelligence, creativity, and wisdom. Journal of Personality and Social Psychology, 49: 607-627. Stets, J. E., & Burke, P. J. 2000. Identity theory and social identity theory. Social Psychology Quarterly, 63: 224–237. Stinchcombe, A. 1965. Social structure and organizations. In James G. March (Ed.), Handbook of Organizations: 142-169. Chicago: Rand McNally. Stuart, T. E., Hoang, H. & Hybels, R. C. 1999. “Interorganizational Endorsements and the Performance of Entrepreneurial Ventures”, Administrative Science Quarterly, 44: 315‐349. Sørensen, J. B. 2007. Bureaucracy and entrepreneurship. Administrative Science Quarterly, 52: 387-412. Sørensen, J. B., & Sorenson, O. 2003. From conception to birth: Opportunity perception and resource mobilization in entrepreneurship. Advances in Strategic Management, 20: 89-117 Tajfel, H. 1972. Social categorization (English translation of "La categorisation sociale"). In S. Moscovici (Ed.), Introduction a la psychologie sociale: 272-302. Paris: Larousse. 135 Tajfel, H., & Turner, J. C. 1979. An integrative theory of intergroup conflict. In W. G. Austin & S. Worchel (Eds.), The social psychology of intergroup relations: 33– 47. Monterey, CA: Brooks-Cole. Taylor, T. W. 1996. Earnings, independence or unemployment: Why become self-employed. Oxford Bulletin of Economics and Statistics, 58: 253-266. Turner, J. C. 1968. The self-conception in social interaction. In C. Gorden & K. L. Gergen (Eds.), The self in social interaction: 93–106. New York: Wiley. Turner, J. C. 1985. Social categorization and the self-concept: A social cognitive theory of group behavior. In E. J. Lawler (Ed.), Advances in group processes: Theory and research: 77- 122. Greenwich, CT: JAI Press. Van Stel, A., Carree, M., & Thurik, R. 2005. The effect of entrepreneurial activity on national economic growth. Small Business Economics, 24: 311-321. von Hippel, E. 1986. Lead users: a source of novel product concepts. Management Science, 32: 791-805. von Hippel, E. 2005. The democratization of innovation. Cambridge, Mass. Hippel, E. V., & Krogh, G. V. 2003. Open source software and the “private-collective” innovation model: Issues for organization science. Organization Science, 14: 209-223. Wann, D. L., Hamlet, M. A., Wilson, T. M., & Hodges, J. A. 1995. Basking in reflected glory, cutting off reflected failure, and cutting off future failure: The importance of group identification. Social Behavior and Personality, 23: 377- 388. Ward, T. B. 2004. Cognition, creativity, and entrepreneurship. Journal of Business Venturing, 19: 173-188. Weber, C. Johnson, M., & Arceneaux, K. 2011 Genetics, personality, and group identity. Social Science Quarterly, 92: 1314-1337. Weisbrod, B.A. 2000. To profit or not to profit: The commercial transformation of the nonprofit sector. Cambridge University Press: Cambridge, U.K. Wennberg, K. 2009. Entrepreneurial exit. Doctoral dissertation, Linköping. Wennberg, K., Wiklund, J., DeTienne, D. R., & Cardon, M. S. 2010. Reconceptualizing entrepreneurial exit: Divergent exit routes and their drivers. Journal of Business Venturing, 25: 361-375. 136 Whetten, D. A., & Mackey, A. 2002. A social actor conception of organizational identity and its implications for the study of organizational reputation. Business & Society, 4: 393–414. Yang, T., & Aldrich, H. E. 2012. Out of sight but not out of mind: Why failure to account for left truncation biases research on failure rates. Journal of Business Venturing, 27: 477-492. Zhang, A., Arvey, R., 2009. Rule Breaking in Adolescence and Entrepreneurial Status: An Empirical Investigation. Journal of Business Venturing, 24: 436–447. Zhao, H., & Seibert, S.E. 2006. Big five personality dimensions and entrepreneurial status: A metaanalytical review. Journal of Applied Psychology, 91: 259-271. Ziedonis, R. H. 2004. Don't fence me in: Fragmented markets for technology and the patent acquisition strategies of firms. Management Science, 50: 804-820 137 TABLE 1 Dimensions of Identity and Variance in Meanings for Mavericks, Missionaries and Masterminds Maverick Missionary Mastermind Identity dimension Basic social motivation Self-interest and risk Promote ideas and pursue visions Mutual support with the community Survey item I am someone who takes chances I am someone who builds castles in the sky I hack because I think it is important to solve problems/bugs or add new features Basis of self-evaluation Refusing to be bound by, accepted beliefs, customs, or practices. Concern to responsibly act upon social wrongs Learning to be able to bring benefits to the community Survey item I am someone who is a nonconformist I am someone who questions social norms, truisms, assumptions I hack because I want to learn new things Frame of reference Competition Society Community Survey item I hack because I dislike proprietary products and I want to defeat them I hack because I feel a personal obligation to contribute to the world I hack because I feel a personal obligation to contribute to my hacker community Variance in meanings 137 TABLE 1 Dimensions of Identity and Variance in Meanings for Mavericks, Missionaries and Masterminds Maverick Missionary Mastermind Identity dimension Basic social motivation Self-interest and risk Promote ideas and pursue visions Mutual support with the community Survey item I am someone who takes chances I am someone who builds castles in the sky I hack because I think it is important to solve problems/bugs or add new features Basis of self-evaluation Refusing to be bound by, accepted beliefs, customs, or practices. Concern to responsibly act upon social wrongs Learning to be able to bring benefits to the community Survey item I am someone who is a nonconformist I am someone who questions social norms, truisms, assumptions I hack because I want to learn new things Frame of reference Competition Society Community Survey item I hack because I dislike proprietary products and I want to defeat them I hack because I feel a personal obligation to contribute to the world I hack because I feel a personal obligation to contribute to my hacker community Variance in meanings 138 TABLE 2 Descriptives and Factor Loadings for Mavericks, Missionaries and Masterminds Questions Mean STD Factor Uniqueness Variance Proportion Cronbach's alpha I am someone who takes chances 5.27 1.22 0.61 0.63 I am someone who is a nonconformist 5.12 1.38 0.81 0.34 I hack because I dislike proprietary products and I want to defeat them 4.64 1.80 0.62 0.62 I am someone who builds castles in the sky 4.76 1.53 0.66 0.56 I am someone who questions social norms, truisms, assumptions 5.8 1.24 0.75 0.44 I hack because I feel a personal obligation to contribute to the world 5.35 1.44 0.66 0.56 I hack because I think it is important to solve problems/bugs or add new features 5.68 1.24 0.76 0.42 I hack because I want to learn new things 6.55 0.80 0.68 0.54 I hack because I feel a personal obligation to contribute to my hacker community 5.02 1.47 0.74 0.45 0.41 0.45 0.54 Rotated factor loadings Mastermind 1.59 0.53 Maverick 1.41 0.47 Missionary 1.44 0.48 138 TABLE 2 Descriptives and Factor Loadings for Mavericks, Missionaries and Masterminds Questions Mean STD Factor Uniqueness Variance Proportion Cronbach's alpha I am someone who takes chances 5.27 1.22 0.61 0.63 I am someone who is a nonconformist 5.12 1.38 0.81 0.34 I hack because I dislike proprietary products and I want to defeat them 4.64 1.80 0.62 0.62 I am someone who builds castles in the sky 4.76 1.53 0.66 0.56 I am someone who questions social norms, truisms, assumptions 5.8 1.24 0.75 0.44 I hack because I feel a personal obligation to contribute to the world 5.35 1.44 0.66 0.56 I hack because I think it is important to solve problems/bugs or add new features 5.68 1.24 0.76 0.42 I hack because I want to learn new things 6.55 0.80 0.68 0.54 I hack because I feel a personal obligation to contribute to my hacker community 5.02 1.47 0.74 0.45 0.41 0.45 0.54 Rotated factor loadings Mastermind 1.59 0.53 Maverick 1.41 0.47 Missionary 1.44 0.48 139 TABLE 3 Descriptive Statistics and Correlation Coefficients for Variables in First Stage, Probit Regression (N=646) Variables Mean STD 1 2 3 4 5 6 7 8 9 10 11 1 Entrepreneur 0.433 0.496 2 Opportunity Implementation 0.313 0.464 0.246 3 Maverick 0.0430 0.997 0.129 0.132 4 Missionary 0.0542 0.980 0.146 0.161 0.514 5 Mastermind 0.091 0.904 -0.038 0.057 0.164 0.362 6 Creativity 0.084 0.933 0.221 0.188 0.353 0.413 0.212 7 Extrinsic Motivation -0.001 1.018 0.107 0.115 0.053 0.179 0.283 0.147 8 Intrinsic Motivation 0.097 0.894 -0.029 0.030 0.180 0.223 0.339 0.220 -0.010 9 Openness to experience 0.067 0.972 0.165 0.103 0.182 0.233 0.069 0.443 -0.043 0.123 10 Agreeableness 0.040 1.002 -0.006 -0.055 0.038 0.175 0.222 0.140 0.076 0.107 -0.059 11 Conscientious-ness 0.001 1.024 0.008 0.065 -0.042 -0.085 0.002 0.028 0.042 -0.075 -0.003 0.001 12 Growth mind 0.000 1 0.139 0.039 0.209 0.218 0.172 0.204 -0.040 0.203 0.236 0.145 0.017 13 Goal oriented 0.759 0.428 0.034 0.014 0.083 0.173 0.174 0.168 0.182 0.108 0.087 0.101 0.144 14 Reputation 0.560 0.497 0.007 0.079 0.053 0.153 0.282 0.065 0.427 0.133 -0.064 -0.001 -0.015 15 Contribution 0.920 0.272 0.052 0.077 0.042 0.041 0.086 0.050 -0.001 0.189 0.059 0.047 -0.039 16 Mean centered age -0.346 9.414 0.260 0.120 0.008 0.011 -0.051 0.134 0.017 -0.016 0.141 -0.117 0.005 17 (Mean centered age)2 8.861 1.600 0.098 0.066 -0.022 -0.015 0.017 0.070 0.054 0.028 0.044 -0.093 -0.013 18 Occupation enjoyment 0.639 0.481 0.091 0.117 -0.014 -0.005 0.055 0.036 0.023 0.128 -0.029 0.082 0.049 19 Female 0.230 0.422 -0.049 -0.021 0.020 0.033 0.003 -0.102 0.012 0.016 -0.115 0.049 -0.038 20 Married/ Relationship 0.437 0.496 0.118 0.059 0.018 0.032 -0.021 0.127 -0.052 0.034 0.088 0.001 0.000 21 Children 0.180 0.384 0.144 0.085 0.014 -0.009 -0.013 0.131 0.013 0.045 0.098 -0.073 -0.052 22 Region 0.574 0.495 0.203 0.095 0.058 0.118 0.008 0.300 0.172 0.147 0.153 -0.003 0.018 12 13 14 15 16 17 18 19 20 21 22 12 Growth mind 13 Goal oriented 0.059 14 Reputation -0.088 0.142 15 Contribution 0.023 0.006 -0.010 16 Mean centered age 0.011 0.015 -0.054 -0.0600 17 (Mean centered age)2 -0.044 0.069 -0.019 -0.003 0.594 18 Occupation enjoyment 0.083 0.096 0.023 -0.009 -0.006 0.016 19 Female -0.050 -0.052 0.063 -0.095 -0.071 -0.149 -0.063 20 Married/ Relationship 0.047 0.081 -0.038 -0.015 0.286 0.138 0.076 -0.171 21 Children 0.028 0.076 0.024 0.020 0.502 0.310 0.041 -0.170 0.426 22 Region 0.137 0.107 0.089 0.137 0.220 0.093 -0.008 -0.064 0.108 0.183 140 TABLE 4 Descriptive Statistics and Correlation Coefficients for Variables in Second Stage, Logistic Regression (N=547) Variables Mean STD 1 2 3 4 5 6 7 8 9 10 11 12 1 Exit 0.080 0.272 2 Maverick 0.181 0.920 -0.047 3 Missionary 0.100 0.932 -0.018 0.588 4 Mastermind -0.007 0.910 -0.075 0.195 0.205 5 Creativity 0.298 0.922 -0.063 0.363 0.394 0.342 6 Extrinsic Motivation 0.046 0.950 -0.050 0.024 0.165 0.227 0.060 7 Intrinsic Motivation 0.066 0.944 -0.011 0.047 0.217 0.417 0.278 0.012 8 Openness to experience 0.290 0.845 -0.013 0.252 0.328 0.150 0.412 0.025 0.179 9 Agreeableness -0.027 0.993 -0.060 0.105 0.225 0.151 0.209 0.095 0.228 0.035 10 Conscientiousness 0.092 1.009 -0.064 -0.042 -0.131 -0.060 0.016 0.003 -0.089 0.001 -0.029 11 Growth mind 0.135 0.979 -0.021 0.089 0.065 0.176 0.151 -0.053 0.275 0.242 0.213 0.072 12 Goal oriented 0.722 0.448 -0.012 0.129 0.161 0.153 0.183 0.184 0.119 0.168 0.099 0.198 0.094 13 Reputation 0.492 0.500 -0.022 -0.043 0.037 0.252 0.035 0.370 0.226 -0.089 0.132 -0.043 -0.028 0.014 14 Contribution 0.940 0.238 -0.010 0.003 -0.014 0.070 0.077 -0.132 0.206 0.055 0.032 -0.001 0.103 0.066 15 Mean centered age 5.166 1.027 -0.056 -0.019 -0.009 0.075 0.162 -0.064 0.091 0.165 -0.099 -0.063 0.010 0.016 16 (Mean centered age)2 1 320 2.360 -0.079 -0.061 -0.021 0.041 0.105 -0.034 0.102 0.120 -0.156 -0.053 -0.092 0.012 17 Occupation enjoyment 0.700 0.459 -0.056 -0.076 0.100 0.148 0.073 0.067 0.149 0.044 0.233 0.019 0.102 0.102 18 Female 0.203 0.403 0.001 -0.038 -0.0830 -0.010 -0.055 0.092 -0.011 -0.155 0.027 0.017 -0.102 -0.012 19 Married/Relationship 0.537 0.499 0.018 -0.098 -0.034 -0.001 0.107 0.095 0.080 0.082 -0.045 -0.003 -0.005 0.1450 20 Children 0.293 0.455 -0.028 -0.077 -0.028 0.004 0.048 0.002 0.064 0.070 -0.056 -0.191 -0.110 0.040 21 Region 0.678 0.468 0.017 0.017 0.134 0.113 0.348 0.207 0.136 0.117 0.160 0.049 0.183 0.228 22 Disrupt 0.064 0.244 0.115 0.145 0.045 0.029 -0.048 -0.088 -0.072 -0.083 -0.081 -0.031 0.050 0.112 23 IT industry 0.665 0.472 0.053 -0.174 -0.237 0.032 -0.189 -0.163 0.113 -0.006 -0.205 0.029 0.072 -0.059 24 Firm size 1.470 2.743 -0.024 -0.029 0.010 0.010 0.033 -0.057 -0.135 -0.145 -0.135 0.168 -0.213 0.097 25 Team size 1.408 2.419 -0.019 0.005 0.103 0.062 0.077 -0.079 -0.023 -0.057 -0.090 0.080 -0.100 0.091 26 Invmills 0.720 0.336 0.094 -0.245 -0.379 -0.003 -0.4893 -0.218 0.026 -0.440 -0.180 -0.021 -0.303 -0.095 13 14 15 16 17 18 19 20 21 22 23 24 25 26 13 Reputation 14 Contribution -0.027 15 Mean centered age -0.104 -0.009 16 (Mean centered age)2 -0.070 0.028 0.836 17 Occupation enjoyment 0.237 -0.116 -0.066 -0.005 18 Female 0.104 -0.101 -0.127 -0.130 -0.027 19 Married/Relationship 0.018 -0.0350 0.268 0.196 0.081 -0.270 20 Children 0.0830 -0.040 0.550 0.407 0.035 -0.205 0.403 21 Region -0.004 0.072 0.139 0.064 -0.024 -0.032 0.177 0.133 22 Disrupt -0.018 -0.060 -0.007 -0.087 0.041 -0.132 0.153 0.128 0.004 23 IT industry 0.031 0.195 -0.101 -0.025 0.094 -0.008 -0.013 -0.047 -0.289 -0.052 24 Firm size -0.023 0.029 0.029 0.020 -0.040 0.016 0.132 0.051 0.082 0.277 0.009 25 Team size 0.013 0.046 -0.090 -0.060 0.016 -0.025 0.049 -0.045 0.087 0.194 0.060 0.645 26 Invmills -0.006 -0.044 -0.499 -0.313 -0.173 0.103 -0.214 -0.232 -0.476 0.039 0.259 0.104 0.091 Table 4 Descriptive Statistics and Correlation Coefficients for Variables in Second Stage, Logistic Regression (N=547) 141 TABLE 5 Determinants of hazards of transition to entrepreneurship and firm exit Model 1 Model 2 Model 3 Model 4 Variables Probit - Entrepreneur Logit - Exit Logit - Exit Logit - Exit (3years) Opportunity Implementation 0.488*** (4.18) Maverick 0.052 -0.392 (0.82) (-1.05) Missionary 0.092 0.801* (1.29) (2.46) Mastermind -0.173* -0.896** (-2.32) (-3.01) Creativity 0.137+ -0.0659 -0.0708 0.377 (1.79) (-0.35) (-0.32) (1.36) Extrinsic Motivation 0.124+ -0.140 -0.204 0.0740 (1.92) (-0.65) (-0.79) (0.31) Intrinsic Motivation -0.150* 0.0129 -0.0825 -0.327 (-2.19) (0.06) (-0.34) (-1.10) Openness to experience 0.067 0.205 0.295 0.379+ (0.98) (0.89) (1.26) (1.65) Agreeableness 0.009 -0.207 -0.155 -0.140 (0.17) (-1.08) (-0.79) (-0.75) Conscientiousness -0.020 -0.359+ -0.316 -0.374+ (-0.36) (-1.84) (-1.62) (-1.69) Growth mind 0.144* 0.00573 -0.182 0.0885 (2.33) (0.03) (-0.85) (0.37) Goal oriented -0.035 0.156 -0.0781 -0.246 (-0.26) (0.38) (-0.17) (-0.52) Reputation -0.014 0.174 0.300 0.617 (-0.11) (0.46) (0.73) (1.38) Contribution 0.330+ -0.124 -0.333 0.0392 (1.69) (-0.18) (-0.42) (0.05) Mean centered age 0.037*** 0.0728 0.0852 0.195** (4.33) (1.41) (1.53) (3.13) (Mean centered age)2 -0.001 -0.00561* -0.00532* -0.00815** (-1.41) (-2.16) (-2.04) (-2.77) Occupation enjoyment 0.261* -0.316 -0.240 -0.0556 (2.24) (-0.80) (-0.55) (-0.12) Female -0.012 -0.0689 0.0289 0.0417 (-0.09) (-0.16) (0.07) (0.09) Married/relationship 0.081 0.408 0.277 0.411 (0.67) (1.13) (0.70) (0.91) Children -0.076 -0.487 -0.759 -1.372* (-0.44) (-0.87) (-1.45) (-2.40) Region 0.249* 1.236* 1.659* (2.10) (2.08) (2.45) year 2 2.107** 2.199** 2.294** (2.73) (2.83) (2.86) year 3 2.399** 2.579*** 2.789*** (3.10) (3.31) (3.45) year 4 2.403** 2.622*** 2.927*** (3.06) (3.31) (3.49) Disrupt 1.510* 1.815** (2.44) (2.85) IT Industry 0.638 0.862+ (1.30) (1.81) Firm size -0.144 -0.121 (-1.33) (-1.26) Team size -0.0174 -0.0702 (-0.19) (-0.78) invmills 1.143 1.965+ 5.053** (1.39) (1.84) (3.03) Constant -0.878*** -5.089*** -6.800*** -10.66*** (-3.63) (-3.86) (-3.89) (-4.15) Number of observations 646 547 547 547 Log-likelihood -374.640 -134.414 -127.414 -119.805 Wald chi2 116.05 33.08 52.52 57.04 Pseudo R2 0.153 0.1219 0.1676 0.217 + p<0.1, * p<0.05, ** p<0.01, *** p<0.001; t-statistics in parentheses