How could one determine if the distribution of sample means is approximately normal?
I. Both np and n ($1$-p) are greater than $10$, where n is the number of trials and p is the proportion of success.
II. The sample is large, say greater than $30$.
III. The parent population is approximately normal.