The quality of the original data in a study is the single most important parameter in determining the value of the study. If low quality data are obtained in a scientific investigation, even the most sophisticated software cannot transform it into valuable, reliable, actionable interpretations. In this chapter the attributes of high quality data including technique and reagent validation, reproducibility, positive and negative controls, operative principles for the collection of samples, the propriety of study models, and the correspondence of data collected with interpretations desired are described. Explicit validation is introduced as an imperative for analytical approaches that are essentially opaque such as artificial intelligence and predictive modeling. The inverse relationship between quality and the amount of data collected and the complexity of the scientific process are also considered.