Bias

In statistics, bias is a systematic error that pushes results consistently in one direction, as opposed to random error that scatters them harmlessly around the truth. A biased bathroom scale that always reads 2 kg heavy is wrong every single time, in the same way.

Bias is dangerous precisely because more data does not fix it — you just get a very precise wrong answer. It can creep in through how you collect data (a survey only of internet users), how you measure (a leading question), or how you analyse (quietly dropping inconvenient cases). The cure is not volume but design: noticing the systematic tilt before it bakes itself into your conclusions.