Building Your Data Field Guide

How to go beyond just data dictionaries

When encountering a new dataset, a data dictionary is helpful for understanding and exploring it. Data dictionaries are typically the first tool for EDA, but they still lack or even obscure context. Understanding the data is required before any models or analyses can be built. For datasets with 50 or more columns, constantly looking up information in the dictionary is a mental tax that slows …

building data data science editors pick exploratory-data-analysis intuition watercolor