Barry Eggleston is a health research statistician who has worked on both clinical trials and observational studies, and is currently with RTI in North Carolina. In his early career, his work was solely designing and analyzing clinical trials using typical biostatistics methods ranging from t-test to survival analysis and mixed models. After moving to RTI International in 2010, he began to apply Bayesian statistics in clinical trial analyses, use propensity scoring for analysis of observational data, and learn machine learning methods for predictive modeling. Since 2010 his awareness of the world of data science has increased as well as his perception of where things could go in the future.
Several years ago, he would go to a statistics conference and some sections were on statistics and while others were data science type sections. By 2019, he would go to a conference and traditional statistics and data science seemed to be very intermingled. Even the presidential address talked about statistics and data science together and how we as statisticians needed to work with data scientists to help them understand the effects of uncertainty on their results. Which solution is best will be determined by project goals and not any statistics vs. data science tradition exposure.
Barry needs to integrate the methods of traditional statistics and data science himself in his own work, “because if I don’t I may not be able to collaborate with future colleagues who have both data science and statistical methods so well integrated into their work.”
Having taken a number of Statistics.com courses previously, Barry came back to Statistics.com to take Predictive Analytics using Python, saying this would enable him to
“communicate better with the data science group at work since they are mainly a python shop.”