AOM 2021 Annual Meeting -- PDW:
The rise of computational methods in theory building

As digital traces and computational capacity has grown exponentially, researchers in the IS field, as well as adjacent fields, have increasingly utilized computational methods of various kinds. Recently, computational methods have also been used in an inductive mode, geared towards developing, rather than testing, theory. In this PDW we seek to examine the emerging role of computational methods in theory development, associated potentialities as well as obstacles to overcome. We do so through a number of presentations of empirical work currently underway, group discussions, as well as a panel of senior scholars. This workshop is intended to contribute to the increasing sophistication, prevalence, and legitimacy of computational methods in developing new exciting theories of information systems and organizing.
AOM 2021 Annual Meeting -- PDW:
Integrating Induction and Abduction for Theory Development in Big Data Research

The age of big data has brought the “problem of plenty” whereby researchers have to grapple with an abundance of observations to consider in theory development. Advancements in artificial intelligence (AI) and machine learning (ML) are enabling the inductive discovery of patterns and trends in the big data. Yet, developing generalizable theory from the abundance of observations and patterns seems elusive. In the third decade of the 21st century, it is ever important to develop theory that is sensitive to the diversity – in contexts and geographies. Responding to the Bringing the Manager back in Management Theme, we invite scholars to integrate perspectives of induction and abduction for developing theory that is actionable for managers in the 21st century. Deduction, induction and abduction denote three distinct approaches of knowledge creation. Patterns discovered in big data via induction can serve as inputs for abduction which is inference to the best explanation. This PDW addresses theory development challenges of transparency and explainability, ethics and inclusivity, and cognitive reapportionment across system and human cognitors while integrating induction and abduction. Research on digital innovation, platforms and health care, and socio-technology design contexts provide exemplars of this approach, including demonstrations of induction via ML techniques and induction/abduction in qualitative and ethnographic studies. Our interactive workshop format is designed to engage academics and data scientists in addressing techniques, trade-offs and ethical issues for theory development in big data research.