If the Street Is Wet, Is It Raining? What the Fallacy of Affirming the Consequent Reveals about the Misuse of Valid Statistical Evidence

To register please email business.infosystems@sydney.edu.au.

Abstract

The idea lies at the intersection of formal logic, empirical science, and statistical evidence.  In brief, the idea is that empirical scientific research that uses valid statistical evidence has misused it in the form of what formal logic calls “the fallacy of affirming the consequent.”  In the research seminar, I will engage in a “deep dive” into two published articles to illustrate this idea: (1) “User Acceptance of Computer Technology: A Comparison of Two Theoretical Models,” which Fred D. Davis, Richard P. Bagozzi, and Paul R. Warshaw published in Management Science in 1989 and which is notable for having been cited over 38,000 times according to Google Scholar, and (2) “Digital M&A, Digital Innovation, and Firm Performance: an Empirical Investigation,” which André Hanelt, Sebastian Firk, Björn Hildebrandt and Lutz M. Kolbe published in European Journal of Information Systems in 2021 and which is notable for having received one of the Association for Information Systems’ five Best Information Systems Publications Awards for 2021.

Concerns that would benefit from feedback from the seminar participants include: (1) To what extent would a paper written around this idea need to provide documentation of the extent of this phenomenon (the phenomenon of committing the fallacy of affirming the consequent in the use of statistical evidence in published information systems research)? (2) Is qualitative research also susceptible to the claim that it has misused evidence by committing the fallacy of affirming the consequent? (3) Should I write a paper that covers not only the problem (namely, how empirical science has misused statistical evidence), but also the remedy (namely, how empirical science may properly use statistical evidence)?  This would involve introducing “modus tollens” and the notion of falsifiability, along with the disadvantage of introducing the baggage associated with the philosophy of Karl Popper.  (4) What would be the genre of a paper written about this?  It would not be theory building or theory testing.  And what would the literature review cover?  (5) Would partnering with co-authors be helpful in addressing these concerns?