May, 21th 2024

Tuesday

Manuel Wiesche

(TU Dortmund)

Title: Bad Client Feedback on Digital Labor Platforms: How Freelancers Deal with the Peril Posed by Negative Reviews on Upwork

Abstract: Freelancing on digital labor platforms is an important source of income for gig workers. As self-employed professionals, freelancers rely on positive reviews to attract new jobs and new clients. It is well-known that positive reviews help to attract clients by reducing uncertainty about freelancers' ability to deliver services; however, we do not know how negative reviews affect freelancers' success nor what strategies freelancers can use to deal with negative reviews. This is especially important because we know that negative reviews on e-commerce platforms undermine product sales. To identify strategies that freelancers use to overcome negative reviews, we conducted a fuzzy set qualitative comparative analysis (fsQCA) of more than 1,500 freelancer profiles. Our analysis reveals that sets of communication skills, collaboration skills, and sharing complete information influence freelancers' success. In the presence of negative ratings, we identify five pathways and six conditions that predict freelancer success on digital labor platforms. We conclude with implications for research and practice.

June, 04th 2024

Tuesday

Veeresh Thummadi

(University of Galway)

Title: Moving the streetlight of information systems research through computational social sciences approaches

Abstract: Ever-increasing volumes of trace data eventually threaten to exceed the limits of human cognition and manual processing for studying information systems phenomena. The current state of IS research calls for significant methodological innovations through computational approaches that use the cognition of both humans and artificial intelligent agents. In line with this, I argue that IS researchers need to embrace computational social science approaches for studying IS phenomena and utilize trace data and analytical techniques for theorizing. In this talk, I discuss how computational approaches have been utilized for studying three IS phenomena namely 1) agile and waterfall design methods 2) big data software development and 3) Systematic Literature Reviews (SLRs) of DevOps.

June, 11th 2024

Tuesday

Julian Lehmann

(Arizona State University)

Title: Addressing Adverse Generativity to Sustain Legitimacy: A Field Study of the Emergence of an Esports Platform

Abstract: Platform ventures continually innovate, and this often results in unintended, adverse behavior on the platform – what we refer to as “adverse generativity.” Platform ventures must address this adverse generativity to build and maintain legitimacy. To explore how platform ventures sustain pragmatic legitimacy in the face of adverse generativity, we report on a case study of a leading esports platform over an 18-year period and how it dealt with increasingly inventive forms of cheating on the platform. Our study suggests that adverse generativity evolves in an arms race, of sorts, and addressing it requires a cumulative effort of sociotechnical actions. We develop a process model of how platform ventures sustain pragmatic legitimacy in the face of adverse generativity by continuously iterating through containment, deterrence, and engagement, and we discuss implications for the literature on digital platforms and organizational legitimacy.

June, 24th 2024

Monday

Kai Lung Hui

(HKUST)

Title: Sentiment influences on sentiment with discussion in football subreddits

Abstract: We investigate sentiment contagion in social media. We find strong evidence of the presence of sentiment contagion in social media after controlling for concurrent events and user’s past behavior. The contagion is robust within the same thread as well as across different threads of the same forum. However, the impact of the treatment content varies depending on the source of thread when the consecutive messages are posted in different threads. We find weak empirical evidence that more experienced social media users may exhibit lower sensitivity to such contagion and the presence of negativity bias. We then examine how the sentiment of the exposed contents influences the future behaviors of social media users. We find that the more users are exposed to extreme content, either negative or positive, the less likely they are active in both the short- and long-term, including even two years after the treatment period. We draw related managerial and practical implications.

July, 2nd 2024

Tuesday

Kevin Bauer

(Uni Mannheim)

Title: Please take over: XAI, delegation, and domain knowledge

July, 9th 2024

Tuesday

Christoph Wolff

(Smart Freight Centre, Amsterdam)

Title and Abstract will follow shortly!