>[!abstract] >Simpson’s paradox is a statistical phenomenon in which a trend that appears in several separate groups reverses or disappears when the data are combined. It occurs when an unaccounted confounding variable influences the relationship between the variables being studied, causing aggregated results to misrepresent subgroup realities. For example, a treatment might appear effective overall but ineffective within each demographic subgroup once underlying differences are considered. The paradox illustrates the importance of context, stratification, and causal reasoning in data analysis, showing that correlations can invert depending on how data are grouped or interpreted. >[!related] >- **North** (upstream): [[Veridical paradox]] >- **West** (similar): — >- **East** (different): — >- **South** (downstream): —