Bias

A systematic error, which contributes to the difference between a population mean of measurements or test results and an accepted reference value


Bias

A systematic error, which contributes to the difference between a population mean of measurements or test results and an accepted reference value

Použito v metodice