|
|
|
|
front |1 |2 |3 |4 |5 |6 |7 |8 |9 |10 |11 |12 |13 |14 |15 |16 |17 |18 |review |
 |
The sample size applies to the comparison we want to make. If it is important to have an estimate of prevalence for each strata and sex in which we will classify our subjects, then the sample size calculated apply to each of these estimations. Otherwise the over all estimate will be within the bounds prescribed in the design but the individual comparisons will lack both precision and power; rendering the comparisons virtually useless. |