use the error bound for the integral test to determine an upper bound on the error using the partial sum sn