|
|
|
|
front |1 |2 |3 |4 |5 |6 |7 |8 |9 |10 |11 |12 |13 |14 |15 |16 |17 |18 |review |
 |
We must have an idea of the prevalence from another studies or just guess. It is better to err closer to 50% so the estimate will be within the bounds desired. A prevalence of 50% needs the largest sample size. |