Question 1. What additional information would you need? Post your information request in the comments below. (Additional information needed will be posted on Wednesday of Week 3.)

Needed programmatic threshold. Coverage accepted by the Government to calculate Delta to classify provinces / districts , additionally needed alpha (precision ) and beta (CI)

Question 2. Based on the new request, what would be the overall sample size?

For estimating coverage of 10 provinces:

ESS per strata = 264

The overall sample size would be for 10 strata 10,600

Given

Confidence interval ±5% of each Stratum,

Existing coverage estimate 85%,

Design effect 4 and 10 respondents per cluster.

Total households would be visited 62,540 ( Five (5) houses to find one eligible respondent @ 15% non-response )

For classifying provincial coverage:

The steering committee wants to know if any province coverage estimates below 90% should be classified as fail

Programmatic threshold

Delta =10%

Direction Below

Alpha=5%

Beta=20%

Given the parameters values as above ESS=272 and overall sample size would be 10,880

Question 3. What decision would you make regarding the survey scope and objectives? What is your rationale for this decision?

As desired by the committee , the above sample size, both for classification and estimation give us the smaller sample size (for 10 provinces as compared to 100 districts) and also expecting to get the reliable estimates at provincial level. However the inter and intra district disparities in vaccine coverage would not be detected as the average at provincial level will mask the inequities with in the province and districts. But the sample size is feasible and manageable to ensure good quality survey. Good reliable estimates at the provincial level will help the Government in terms of planning and prioritization process to accelerate the activities in failing provinces. As we learn from the Vaccination Coverage Survey manual, “it will rarely be a cost-effective use of resources to attempt to conduct surveys in every district of a country. At the most peripheral health system levels, practical field methods such as health facility-based assessments can evaluate multiple aspects of service provision, coverage and timeliness of each vaccine among clinic attendees, and can stimulate improvement of vaccination as well as recording practices”. This aspect needs to be discussed with the Steering committee.

 

Question 1. What additional information would you need? Post your information request in the comments below. (Additional information needed will be posted on Wednesday of Week 3.)

Needed programmatic threshold. Coverage accepted by the Government to calculate Delta to classify provinces / districts , additionally needed alpha (precision ) and beta (CI)

Question 2. Based on this new request, what would be the overall sample size?

For estimating coverage of 10 provinces:

ESS per strata = 264

The overall sample size would be for 10 strata 10,600

Given

Confidence interval ±5% of each Stratum,

Existing coverage estimate 85%,

Design effect 4 and 10 respondents per cluster.

Total households would be visited 62,540 ( Five (5) houses to find one eligible respondent @ 15% non-response )

For classifying provincial coverage:

The steering committee wants to know if any province coverage estimates below 90% should be classified as fail

Programmatic threshold

Delta =10%

Direction Below

Alpha=5%

Beta=20%

Given the parameters values as above ESS=272 and overall sample size would be 10,880

Question 3. What decision would you make regarding the survey scope and objectives? What is your rationale for this decision?

As desired by the committee , the above sample size, both for classification and estimation give us the smaller sample size (for 10 provinces as compared to 100 districts) and also expecting to get the reliable estimates at provincial level. However the inter and intra district disparities in vaccine coverage would not be detected as the average at provincial level will mask the inequities with in the province and districts. But the sample size is feasible and manageable to ensure good quality survey. Good reliable estimates at the provincial level will help the Government in terms of planning and prioritization process to accelerate the activities in failing provinces. As we learn from the Vaccination Coverage Survey manual, “it will rarely be a cost-effective use of resources to attempt to conduct surveys in every district of a country. At the most peripheral health system levels, practical field methods such as health facility-based assessments can evaluate multiple aspects of service provision, coverage and timeliness of each vaccine among clinic attendees, and can stimulate improvement of vaccination as well as recording practices”. This aspect needs to be discussed with the Steering committee.