Online Program Home
My Program

Abstract Details

Activity Number: 6 - Advocating, Implementing and Explaining Bayesian Analyzes in Statistical Consultations
Type: Invited
Date/Time: Sunday, July 28, 2019 : 2:00 PM to 3:50 PM
Sponsor: Section on Statistical Consulting
Abstract #300147
Title: The Primordial Soup for Bayesian Analysis in Collaborative Settings: Technical Skill, Communication, and Trust
Author(s): Christopher Franck*
Companies: Virginia Tech
Keywords: consulting; collaboration; communication; Bayesian ; bioinformatics; behavioral economics

Even though access to Bayesian methods is growing rapidly, the analytic expectation in many consulting contexts remains frequentist by default. In areas where the frequentist approach is entrenched there are two forces working against the adoption of Bayesian methods in the minds of domain experts. First, domain experts may prefer already published analytic strategies to allow “apples-to-apples” comparisons with current work. Domain experts are wary if novelty in results is partially attributed to domain innovation and partially due to the “new” Bayesian analysis. Second, domain experts worry that their audience (including journal reviewers, sponsors, and other stakeholders) will be less familiar with the interpretation of Bayesian inference, which will muddy waters of written communication and thus dampen impact. The suggestion to use Bayesian methods can hang uneasily unless the statistician’s case for Bayesian methods overcomes these valid concerns. In this talk we will discuss how technical ability, communication, and trust facilitate moving collaborative projects towards Bayesian analysis using bioinformatic and behavioral economic case studies.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2019 program