Data analysis is a fundamentally subjective science. The journey that starts with problem formulation and ends with a set of “final” polished results involves navigating sequences of forks in the data analysis road, each arising from the set of alternative judgment calls that we might have made. However, we rarely stop to consider how these judgment calls affect our results. Have you ever made a seemingly “minor” modification to your analysis and watched your conclusion crumble before your eyes? In this talk, I will shine a spotlight on the role that judgment calls play in data analysis, and will discuss techniques for incorporating these perspectives into data science practice and training. This talk is based on the book "Veridical Data Science: The Practice of Responsible Data Analysis and Decision Making", that I am co-authoring with Professor Bin Yu, the pioneer of the Veridical data science framework that is built around the three guiding principles of data science: Predictability, Computability and Stability (PCS) which unify and expand on ideas and best-practices from both statistics and machine learning.