Developing a Bayesian workshop for full-time staff statisticians.

Area of Science:

• Biostatistics
• Statistical Inference
• Computational Statistics

Background:

• Frequentist and Bayesian statistical inference methods differ in their use of prior knowledge.
• Bayesian methods, though historically significant, faced challenges in adoption due to computational limitations and theoretical debates.
• Increasing computational power and a focus on incorporating prior knowledge have spurred renewed interest in Bayesian approaches.

Purpose of the Study:

• To address the gap in Bayesian methodology skills among biostatisticians trained primarily in frequentist approaches.
• To develop and evaluate a practical, accessible training program for applying Bayesian techniques.

Main Methods:

• A customized, project-based workshop series was designed for full-time biostatistical staff.
• The training emphasized immersive, hands-on learning tailored to accommodate work schedules.
• Program impact was assessed through participant surveys and evaluation of capstone projects.

Main Results:

• All 20 participants successfully completed the workshop.
• Survey results indicated a significant increase in participants' understanding of Bayesian theory and confidence in applying these methods.
• Capstone projects confirmed participants' ability to implement Bayesian methodology.

Conclusions:

• The workshop effectively enhanced the practical skills and confidence of biostatisticians in Bayesian methods.
• The project-based, schedule-accommodating format facilitated full participation and successful skill acquisition.
• Participants are now better equipped to apply Bayesian techniques in their professional work.