dispersion measures for gips

do i have this correct:

Required: 3 year internal dispresion using monthly portfolio returns about the composite aggregate return (choice of standard deviation across equally weighted portfolios or standard deviation with asset weighted composites)

Required iIf management feels standard deviation is inappropriate: an additional 3 year internal dispersion measure other than standard deviation (choice of range of annual returns, high low, interquartile)

Recommended: an external measure, such as standard deviation of composite returns

right/wrong/missing anything? standard 5.A.2 on schweser book 5 page 208 is so unclear.