According to Cassie Kozyrkov, “Data science is the discipline of making data useful.” This includes the sub-disciplines of predictive analytics, artificial intelligence (AI), machine learning and deep learning models, which require one or more datasets with a large number of data points. The private sector drove the creation of "data science" as a discipline, largely as a way to analyze vast amounts of passively generated, unstructured data, such as website analytics. The social sector drove the creation of MERL, which is commonly used to analyze smaller quantities of actively generated data with some structure, such as survey responses. In MERL, data science is used to evaluate impact, understand trends, and inform programmatic decision-making. Communicating information from data science and MERL often comes in the form of visualizations and storytelling.