If the statistical generalization is causally traceable to prior structural injustice, the wrong lies in the structural conditions, not the employer's inference from them—Lippert-Rasmussen's causalargument.
?Rate how convincing each reason is below to see the overall strength.