On serendipity: The happy discovery of unsought knowledge

We may be loath to acknowledge it, but no matter how hard we try, our research designs are not always under our complete control. Some researchers, notably those who conduct laboratory experiments, may take extraordinary measures to control the environmental factors so that they do not influence the key variables under study. But even here, things can go awry. As Lee and Dennis (2012) report in their hermeneutic reinterpretation of the controlled laboratory experiment that was originally reported in Dennis, Hilmer, and Taylor (1998), situational factors unknown to Dennis and his colleagues caused many of the 1998 study's hypotheses to be rejected, in defiance of theoretical explanation. Some methods are less amenable to control, ethnography for instance, where the researcher may abjure any attempt to engage in more than the most rudimentary forms of control. As Descola (1996) reported in his seminal tome “Les Lances du Crépuscule” when ready to set off for the field, his research advisor, the structural anthropologist Lévi-Strauss, suggested that he set aside all his carefully prepared plans and instruments and instead “follow the lie of the land,” adapting his lines of enquiry to the infinite variety of circumstances as they emerged. While most interpretive researchers would not go this far, the idea that they need to be open to “whatever happens” in the world of their research subjects is a very familiar one. For this reason, they have to expect the unexpected. When research is written up for submission, authors of all epistemological persuasions may more or less frequently slip in some of their unexpected, unintended, unanticipated findings. These are often intriguing, lying at the margins of the research design, yet appreciated for their significance. In this vein, Segerdahl, Fields, and Savage-Rumbaugh (2005) explain how research can proceed by accidental discovery, shaping entirely new theoretical conjectures and investigatory possibilities. My own experience conforms to this pattern, since an investigation into knowledge sharing among hotel employees in China led to the identification of the (unexpected and unsought) finding that these same employees were actively working around corporate IT policy, developing their own feral systems to do so (Davison & Ou, 2018). It is in this space that I find inspiration for this editorial, specifically the notion of serendipity, defined by van Andel (1994) as the “art of making an unsought finding” and by the OED as “the faculty of making happy and unexpected discoveries by accident.” 1 I agree with van Andel that serendipity is an art rather than a science: We need to attune our senses to these unexpected phenomena and their affordances for our research, no matter whether they are stark or ethereal in form. In the research context, we need to go beyond mere recognition and develop the confidence to distinguish unexpected findings from their context and then consider how to theorize them. Merton's (1957) view of serendipity was that it involved the “observation of a surprising fact followed by a correct abduction”, i.e., the devising of an explanatory hypothesis or theory. However, not all researchers are equipped with the necessary vision to observe, or for that matter to theorize, these scholarly surprises. As Clarke (2015) remarks, researchers may be blinkered, seeing the world in narrow ways that to a large extent exclude unfamiliar perspectives. I suggest that blinkered researchers are less likely to notice unsought findings, even if they occur, because they are insensitive to knowledge that lies beyond the pale of their world view. We must therefore open our eyes, literally and metaphorically, since, following Heraclitus (Kahn, 1981) (ἐὰν μὴ ἔλπηται μὴ ἔλπηται), if we do not expect the unexpected, we certainly will not find it. I am not (yet) planning a special issue on serendipity in the ISJ, but I do hope that authors will be sensitive to unsought findings in their work and following Merton (1957) to theorize such findings in their submissions to the ISJ where appropriate. In this issue of the Information Systems Journal, we present ten papers. In the first article, Liu, Wang, Min, and Li (2019) draw on the perspective of boundary regulation and dual process theories to assess the consequences of role conflict on social media. A theoretical model is developed to examine the effect of role conflict regarding privacy risk and perceived control, which, in turn, affect self-disclosure, as well as how this process is moderated by high-effort versus low-effort processing. The authors find that users are more likely to follow high-effort processing when habit is relatively low or when emotion is negative. Results from four experiments provide strong support for the model. Their study contributes to the literature by clarifying the effect of role conflict on self-disclosure under high-effort versus low-effort processing. In the second article, Junglas, Goel, Ives, and Harris (2019) investigate how employees, as part of the IT consumerization phenomenon, increasingly use their own devices and choose their own software (e.g., Google Apps, Skype, or Dropbox) in addition to, or instead of, enterprise IT. These employees are thus turning from consumers of enterprise IT to IT deciders, bypassing the IS department to use what critics call “rogue IT.” While discouraged in some contexts, it has been suggested that the influx of consumer IT into the workplace has prompted innovative behaviours among employees. However, until now, there has been no empirical validation of such a beneficial relationship. The authors propose an IT consumerization model and examine the precursors of consumerization and innovative work behaviours. We scrutinize the extent to which an individual's level of satisfaction with enterprise IT in juxtaposition to his perceived level of relative advantage of consumer IT, along with organizational mandates and levels of IT empowerment, influence such behaviours. In the third article, Guhr, Lebek, and Breitner (2019) note that explaining the influence of management leadership on employees' information security behaviour is an important focus in information systems research. Unfortunately, the role of leadership has remained largely unexplored in the information security context. To address this gap in the literature, the authors focus on how the dimensions of the full-range leadership influence employees' intended information security behaviour. Their empirical study takes an interactional psychology perspective and links the dimensions of the full-range model of leadership to employees' security compliance intention and security participation intention. The present study contributes to the literature on information security, management, and leadership by exploring how and why different leadership styles enhance employees' intended information security behaviour. The findings emphasize the importance of transformational leaders because they are capable of directly influencing employees on the extra-role and in-role behaviour levels. The results also indicate new directions for information security and leadership research and implications for leadership practices. In the fourth article, Steininger (2019) reviews 292 articles in IS, entrepreneurship, and general management to conceptualize the roles of IT in entrepreneurial business operations and develop fundamental definitions for the field of IT-associated and digital entrepreneurship research. The findings suggest that IT plays four major roles in entrepreneurship business models: as a facilitator; as a mediator for new ventures' operations; as an outcome of entrepreneurship operations; and as a ubiquity, becoming the business model itself. Leveraging these IT roles, the author then develops a set of definitions to clear up uncertainties surrounding IT-associated new ventures types. For example, he defines digital start-ups as employing business logic that uses entirely IT-mediated processes and solely digital products or services for value creation and transfer. He then outlines a research agenda for IT-associated and digital entrepreneurship research based on identified roles, types and gaps. In the fifth article, Salo, Pirkkalainen, and Koskelainen (2019) examine technostress in the context of social networking sites and services (SNS). To extend prior studies' focus on SNS stressors and use consequences, the authors investigate users' well-being strains and the stressor-triggering technological characteristics of SNS. They analyse narrative interviews of users who have experienced SNS-related stress and reveal four types of well-being strains: concentration problems, sleep problems, identity problems, and social relation problems. The authors also present two explanatory patterns with distinct sets of SNS characteristics and stressors that generate those strains. For instance, self-disclosure features and information cue paucity expose users to life comparison discrepancy, online discussion conflict, and privacy/security uncontrollability stressors, which can lead to identity and social relation problems. The findings can help technostressed users and their stakeholders to identify well-being strains, understand their roots, and avoid them in the future. In the sixth article, Chan, Teoh, Yeow, and Pan (2019) explore how the emergence of disruptive digital innovation (DDI) is redefining how firms achieve competitiveness. Firms have to be agile in order to thrive under such disruptive pressure. While being smaller, less formalized, and organizationally flatter may position Small-and-Medium-Sized Enterprises (SMEs) for greater agility, most SMEs actually suffer from resource constraint as well as narrow and idiosyncratic capabilities that hinder their agility in responding to DDI. The authors present a case study that illustrates how an SME achieves agility in responding to DDI, detailing three key processes of (i) mitigating organizational rigidity, (ii) developing innovative capabilities, and (iii) balancing the tension of organizational ambidexterity. They also demonstrate the importance of incorporating contextual characteristics in generating more nuanced and contextually relevant research insights. In the seventh article, Siering (2019) contributes to the ongoing discussion that integrates information systems and financial markets and focuses on stock touting during Internet-based pump and dump campaigns. Here, manipulators advertise stocks to profit from an increased price level. The paper investigates whether the positive prospects touted actually prevail, whether specific stocks are targeted by manipulators and which factors drive the economic impact of stock touting. The author finds that investors bear the risk to lose substantial amounts of their investments as pump and dump campaigns are followed by large declines in stock prices in days subsequent to the campaigns. The research results are highly relevant for Internet users and market surveillance authorities, as a deep understanding of information-based manipulations is necessary to develop appropriate countermeasures. In the eighth article, Asatiani and Penttinen (2019) seek to gain a deeper understanding of how firms with different degrees of virtuality cope with the discontinuities associated with virtual work. The authors draw on Organizational Discontinuity Theory (ODT) to study discontinuities and continuities constructed in two virtual work environments; a fully virtual firm with no physical office and no regular face-to-face contact between employees; and a hybrid firm with a physical office where employees have the option to work remotely. The authors find that in constructing continuities for virtual work, organizations need to balance rigid and flexible approaches to governance structure, role of technology, communication management, and workflow management. To this end, the authors make four propositions connecting rigidity/flexibility to the degree of virtuality. The authors also reflect on ODT, calling for more research into virtual work continuities and their construction. In the ninth article, Hansen, Gogan, Baxter, and Garfield (2019) conceptualize a patient-present telemedicine consultation as an instance of informed collaboration, which is defined as “ICT-supported collaborative work in which participants … both use system-produced digital information and contribute information to the system.” Reporting on an embedded-cases study of a geriatric telepsychiatry clinic in the northeastern USA, they apply and extend the theory of distributed cognition (DCog). Their study reveals four interrelated dynamic DCog processes, viz., cognitive offloading, visibility of action, intersubjective understanding, and knowledge redundancy, that act as critical enablers of informed collaboration in psychiatry, other forms of health care, and in broader informed collaboration contexts. In the tenth article, Gerlach, Eling, Wessels, and Buxmann (2019) provide a novel perspective on the well-studied topic of information privacy. They examine the challenges associated with our privacy on the Internet faced by the companies who provide the services we use, a significantly under-researched perspective. The paper illustrates the tensions faced by today's companies when they have to balance the need to protect their customers' privacy and, at the same time, to collect and use customer information to improve their services, strategize, and otherwise satisfy their organizational information needs. The authors also identify several tactics that companies use to maintain their balance in this challenging environment. I encourage scholars to build on this study in order to extend our knowledge about information privacy from an organizational perspective. I am grateful to Roger Clarke and Carol Ou for comments received on this editorial.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.