Week 3 Assignment

You are asked to determine what sample size would be needed to obtain an estimate for the national level and classification estimates for the provinces

Answer

The steering committee propose to conduct immunization coverage survey in 10 provinces, if all other parameters remain the same, the sample size calculation results is as follows:

Number of interview to be completed: 10560.

Number of households to visit: 124236 all10 stratum. In HH of each stratum: 12424

Total number of clusters: 1060

Each stratum: 106 clusters.

Each cluster: 118 HH.

The main issue of a survey covering 10 provinces, the steering committee will not obtain granulated information to help determine the coverage survey of each district. However, it is a manageable size and the resource required will be less.

Classification

What additional information would you need? Post your information request in the comments below.

In order to compute the sample size for the classification the following additional parameters is required

Programme threshold: 90% probability to detect a fail if the coverage of DPT is below than 80%

Delta:10%, direction is below

Alfa: 5%

Confidence interval bound: 95%

DPT/Penta 3:

The steering committee propose to conduct immunization coverage survey in 10 provinces, if all other parameters remain the same, the sample size calculation results is as follows:

Number of interview to be completed: 4480.

Total number of households to visit: 52706

HH of each stratum: 5271

Total number of clusters: 450

Number of clusters per stratum: 45.

Number of HH per cluster: 118 HH.

What decision would you make regarding the survey scope and objectives? What is your rationale for this decision?

Scenario one: Estimation

The survey objective is to estimate the coverage of routine immunization with primary focus on DPT /Penta 3 identify

The rational for the steering committee to conduct the survey in 10 provinces is to ensure quality of the survey and to use less resources to undertake the study. The tradeoff is that the steering committee will not obtain a granulated and actionable data

Scenario one: Classification

The survey objective is to determine the level of immunization performance based on the programmatic threshold. The rational for the steering committee to conduct the survey in 10 provinces is operational and quality assurances reasons. To obtain the information needed in in a fewer stratum, they can easily monitor and supervise the survey teams, thus improving quality of data collection. Since they are visiting fewer cluster, they will require fewer survey teams, hence spending less resources. The tradeoff is that the steering committee will obtain data that allows to compare the immunization performance of the provinces. However, they cannot gather a granulated and actionable information.