zum Inhalt springen

Forschungsseminar

Alle Seminare finden im Gremienraum, Pohligstraße 1, 50969 Köln, ab 13.30 Uhr statt, sofern nicht anders angegeben.

Die Vorträge werden als Brown Bag Seminare organisiert, also bieten wir leckere Sandwiches und kalte Getränke an. Alle Interessierten sind herzlich eingeladen, an den Sitzungen teilzunehmen! Wenn Sie Fragen haben, senden Sie bitte eine E-Mail an werder(at)wiso.uni-koeln.de.

Derzeit geplante Seminarvorträge (Referenten und Reihenfolge können sich kurzfristig ändern):

Forschungsseminar Reihe Wintersemester 2018/19
Datum Referent Titel & Zusammenfassung
24.09.2018 Sean Hansen

Sean Hansen

(Rochester Institute of Technology)

The Courage Engine: Distributed Cognition in Agile Development Environments

Traditional software development approaches condition developers to react negatively to late-stage development changes. Agile development contradicts this perspective, encouraging teams to “embrace change,” even late in a project. Agile advocates claim that the network of people, processes, and technologies utilized in these methods provide support to developers by enabling rapid change and a continuous reorientation to the needs of users. In this study, we apply the perspective of distributed cognition to investigate this network with an eye to how agile teams deliver software successfully under conditions of uncertainty. Our findings reveal that agile development environments, including the embedded technologies and artifacts, reflect critical dynamics of distributed cognitive systems. Further, we show that agile teams exhibit the ability to respond to change due to the confidence instilled in the teams via a distributed cognitive system made up of humans, digital technologies, and physical artifacts. This research provides a deeper understanding of the workings of agile teams, establishes a theory-based analysis of the role of the artifacts in socio-technical ecology of agile development, and presents a number of implications for the design of team interactions informed by distributed cognition theory.

15.10.2018 Matthias Sambale

Matthias Sambale

(University of Cologne)

Measuring Fragmentation - A Formal and Empirical Comparison of Fragmentation Metrics

Our consumption of products and information online is mediated largely through software that personalizes the user experience. A possible drawback of personalization is the fragmentation of the user base. There are currently two competing sets of theories and associated metrics for studying fragmentation. The information systems community uses cosine distance to measure the amount of fragmentation in consumption; the literature on development economics and social unrest uses a fragmentation index to study the structure of groups. The two literature streams are not connected, they seem to have evolved independently. It is unclear, whether the two underlying theoretical bases contradict (or not), and if the two metrics would actually measure the same effect of a real intervention. The goal of this paper is to conduct a contest between the two competing theories and metrics. We proof formally that it is indeed possible that the two metrics measure contradictory effects from the same ground truth. We then go on to show that these contradictions do not only occur in artificial examples, but also in “real” empirical data. We conducted a randomized controlled field experiment about the effects of recommender systems on customer behaviour. We establish a causal relationship between the introduction of a recommender system and customer fragmentation. While according to one metric, the recommender system has increased the amount of fragmentation, according to the other metric, fragmentation has gone down. We conclude that the two metrics are actually independent. They seem to measure two different facets of a larger theoretical construct.

12.11.2018 Jan Recker

Jan Recker

(University of Cologne)

Null Hypothesis Significance Testing in Quantitative IS Research: A Call to Reconsider our Practices

Our objective is to promote discussion and policy change amongst Information Systems journals, scholars and students with a vested interest in positivist quantitative research. We focus on knowledge and norms permeating design, analysis and reporting of quantitative studies in information systems, particularly the use of null hypothesis significance testing (NHST) and p-values. Our main argument is that debates about misinterpretations, abuse, and issues with NHST, while having persisted for about half a century, remain largely absent in information systems. We find this an untenable position for a discipline with a proud quantitative tradition. We discuss problems and symptoms associated with the application of NHST and examine how they manifest in current IS research. We review actions currently taken by reference disciplines. To foster a debate, we provide a first set of recommendations for strategic moves that the IS community can make. We do not claim these recommendations will prove ultimately successful; however, they serve to open a debate to jointly identify a strategy for the community.

26.11.2018 Tobias Brandt

Tobias Brandt

(Rotterdam School of Management)

Cascading Effects in Decentralizing Systems: Insights from Vehicle Sharing

Digitization leads to an increasing decentralization of decision-making processes in various key sectors of the economy, such as the energy industry and urban transportation. The sharing of assets enabled by information technology causes every user of those assets to be partially involved in operational decisions regarding the associated system. While the central owner-operator of the system may seek to affect users’ behavior through incentives and nudges, it is often unclear how a particular user’s reaction to such an incentive cascades through the system and affects its overall performance in a stochastic environment. We investigate this question through the example of a free-floating carsharing system, which allows users to end car rentals at any location within the operator’s business area, thereby actively shaping vehicle supply for other users. We analyze how small individual changes in the endpoints of rentals propagate across time and space, influencing overall vehicle utilization rates and revenues. Leveraging a real-world data set of carsharing trips, a discrete-event simulation, and spatio-temporal predictions of vehicle idle times, we illustrate how these minor shifts translate into significant revenue and utilization increases. We also analyze the moderating effect of the operator’s fleet size and a city’s urban structure on these results. Overall our work provides a novel perspective on increasingly decentralized systems that emerge, for instance, as part of the sharing economy and how operators can use incentives and nudges to steer them.

03.12.2018 Derck Koolen

Derck Koolen

(Rotterdam School of Management)

Renewable Technology Non-neutrality and Information Transparency in Decarbonizing Sequential Power Markets

While the influence of information transparency on market efficiency is relatively well investigated from a market point of view, environmental transparency affecting the capability of traders to accurately predict and gather information is still underexposed. Power markets provide us a setting to do so with the ongoing integration volatile renewable energy sources. With different renewable technologies causing information asymmetries between producers and retailers, we propose a multi-stage competitive equilibrium model to analyze the effect on price formation in sequential power markets. We validate the model empirically, analyzing data of sequential power markets in California, the United Kingdom and Germany. The results show evidence for a technology-varying forward premium related to information aggregation of buyers and suppliers. We discuss results and policy measures for creating sustainable smart electricity markets in terms of analytics and IoT devices.

10.12.2018 Peter Gloor

Peter Gloor

(Massachusetts Institute of Technology)

From the Madness of Crowds to the Creativity of Swarms - Leveraging AI for Collaboration and Creativity

Steve Jobs did not create Apple. Steve Jobs created the creative swarm that created Apple. Thanks to the Internet it has never been easier to communicate, collaborate, and innovate on a global scale. At the same time, people on the fringe of society are creating fake news and alternative realities. This talk shows how to positively leverage AI-supported collaboration, where people like Steve Jobs are building creative swarms as Collaborative Innovation Networks (COINs), intrinsically motivated groups of people who work together over the Web, supported by computer technology and AI to create something radically new. Companies like Apple, Google, or Facebook succeed in creating global organizations with collective awareness whose members are united by the same goals and culture to work together towards a shared vision. Their customers form digital tribes which are passionate about their products. Based on over 15 years of research at MIT, this talk illustrates how the creation of these swarms and digital tribes can be supported by augmenting human intelligence in social networks. In collaboration with global fortune 500 firms, our team has identified digital tribes and their preferences on Twitter, predicted the likelihood of senior managers to leave their firm based on their e-mail communication, and measured stress and happiness of individuals in small teams using the “Happimeter”, a smartwatch-based device.

08.01.2019 Oliver Hinz

Oliver Hinz

(Goethe University Frankfurt)

Social Capital Accumulation Through Social Media Networks and Its Benefits in a Project-Based Labor Market: Evidence From a Randomized Online Field Experiment

Work-related social media networks (SMNs) such as LinkedIn introduce novel networking opportunities and features (such as the easy search for potential contacts) that can help individuals establish, extend, and maintain social capital outside of any spatial and temporal barriers. However, extant studies that look at social capital in the context of work-related SMNs tend to focus on the concept’s benefits rather than its dynamics. Thus, those prior studies implicitly assume that network structures are exogenous given to the individuals. Yet, little is known about whether these work-related SMN and their features actually have an impact on the accumulation of digitized social capital. To close the identified gap, we conducted a randomized field experiment whereby we recruited more than 210 freelancers on one large European, freemium-based, work-related SMN. Out of these recruited participants, we randomly assigned 75 freelancers to receive a premium membership subscription voucher, which grant-ed them access to advanced and additional networking features, such as sending private mes-sages to non-contacts. Our results provide the first evidence that individuals will not necessarily accumulate more social capital from being invited to employ those premium features, as the treated freelancers did not automatically change their digitized networking engagement. Instead, those features will only unfold their full utility if the individuals are motivated to proactively as well as purposely engage in strategic networking behavior. To be specific, treated freelancers who sought to use the premium features proactively and purposefully accumulated 4.59% more social capital in the observation period than their counterparts in the control group. We term our finding “theory of purposeful feature utilization”. Also, our dataset allowed us to analyze whether the established amount of digitized social capital affected the freelancer’s real-world revenue stream. We found that younger freelancers who cultivate more digitized social capital seem to generate more monetary revenue from their work-related SMN presence than their older counterparts do.

21.01.2019 Julian Lehmann

Julian Lehmann

(University of Cologne)

The Interaction of Software and Hardware in Digital Product Innovation

Digital product innovation now routinely combines software and hardware components. Voice-enabled home appliances, semi-autonomous robots equipped with sensors and actuators, or 3D printers are all examples of recent digital product innovations that consist of software as well as associated hardware. Generating such digital product innovations requires managing the unique properties of digital technology, as well as their interaction. Thus far however, research on digital innovation lacks an understanding of how software and hardware components interact and how such interactions shape digital product innovation processes. While seminal work on digital innovation has, by and large, focused attention on digital innovation as innovating software components, the interplay between hardware and software domains has profound implications for processes of organizing and coordinating innovation. Yet, most work on digital innovation analytically separates software from hardware, treats hardware as an empirical context for software-based digital innovations, or neglects hardware altogether. In this paper, we argue that understanding digital product innovation depends on the extent to which the interaction between software and hardware components of the underlying digital technologies is appreciated. We present the results of an in-depth case study of Ultimaker – a company that creates and produces 3D printers. We trace the firm’s digital product innovation trajectory over the course of six years (2011-2017) to analyze the interaction between software and hardware components. In a data-driven approach we integrate observations from interviews, archival data, and software development traces. We inductively derive three patterns of interactions in software and hardware components by which digital product innovation advances. Collectively, these patterns increase our understanding of digital innovations in organizations that innovate software and hardware components. We contribute to the literature on digital innovation by (1) giving insight into the interaction of software and hardware components in digital product innovation, and (2) the understanding that hardware components play a more active role than is usually appreciated.

04.02.2019 Andreas Eckhardt

Andreas Eckhardt

(German Graduate School of Management & Law)

The Role of Awareness and Social Relations for Information Security In- and Extra-Role Behavior

18.02.2019 Christian Mahringer

Christian Mahringer

(University of Stuttgart)

Routines and open-ended software development processes – An ethnographic study in Scrum teams

20.02.2019 Johann Kranz

Johann Kranz

(Ludwig Maximilian University of Munich)

To be announced