Confounding

the technical term for what goes wrong when you change multiple variables simultaneously, where a confounding variable correlates with both your change and your outcome, making it impossible to determine which one actually caused the result