If your primary evidence is results-based, why do you assume it's genetics instead of situational? My assumption would be that higher caste students have more

If your primary evidence is results-based, why do you assume it's genetics instead of situational? My assumption would be that higher caste students have more opportunities for proper nutrition, education, etc. Looking at babies adopted into another caste, or people marrying across castes, would help shed some light. But it doesn't seem reasonable that 5k years of caste systems would have such an impact on IQ.