If there are statistical analysis that can answer if the data is biased why isn't the focus of attention all biases rather than just biases that may result in r

If there are statistical analysis that can answer if the data is biased why isn't the focus of attention all biases rather than just biases that may result in racial discrimination? Shouldn't all the arguments be over the data and any biases present (and how to obtain unbiased data) and not the result?