Data mining is the process of using sophisticated statistical techniques to extract and analyze data from large databases to discern patterns and trends that were not previously known. The patterns we find are correlations, but we do not necessarily find causation. Newspaper headlines often link one activity to another and imply causation, rather than a simple correlation or association. Jon Mueller, a professor at North Central College, suggests that when describing a correlation between two variables, there are some acceptable ways to describe this relationship. ACCEPTABLE TERMS for Correlations AVOID THESE TERMS When Discussing Correlations (but they are acceptable when asserting causation) Get Cause Have Increase/decrease Linked Benefits More Impacts More/less, less Enhances/undermines Tied Effect/affect Connected/related Improves/boosts Tend If then type statements (implies one-direction)