I think it'd be a good idea if you defined what you use: what's that sigma function: the sum of all the divisors of the number n? And those p_i's are the prime divisors of n? If so the claim is false:

By the way, without the squares we get equality with n = 6.

Tonio