1. Given this information, what would be the overall sample size?

• 106,000 (spreadsheet: 105,600)

2. How many households would have to be visited?

• 625,400 (spreadsheet: 621,177)

3. What do you think of this estimated sample size?

• Probably the sample size is too large to be practical.

4. How feasible will it be to conduct this survey?

• A survey with this sample size will not be easy. It is financial, logistics and human resource wise will be very demanding. The time needed for complete the survey will also be very lengthy. Therefore, quality of data and final estimate outcome will suffer.

5. What are the trade-offs in terms of time, money, and quality of survey implementation?

• First, the number of stratum can be reduced by aiming the precise estimate at provincial level or limiting only to the national level mainly saving the money and time while estimate precision only to provincial and/or national level
• The Increase of precision level also helps to reduce the sample size considerably but confidence interval will be wider
• The increase of alpha level will reduce the sample size but probability of type I error will be higher.
• Reducing the number of children from one cluster help to reduce the time and money spend on covering one cluster and probably increase the data collection quality