1. Describe the primary sampling units (PSU) and secondary sampling units (SSU) used in country C and in Kinshasa province in the study by Burnett et al.

Ans:

In the case of country C, the Primary Sampling Unit (PSU) was Clusters A cluster was recognized as the Enumeration Areas (EA) with about 200 households.

The Secondary Sampling Units (SSU) for country C were the households in the selected clusters

At the second stage, households will be used as the secondary sampling units (SSU) using

In the Case of Kinshasa Province by Burnett et.al

The Primary Sampling Unit were the Zones which are 12 zones in total.

The neighborhoods were the Secondary Sampling Units (SSU) in which random starting points using ARC GIS ver 10.2 to correspond to the 15 neighborhoods in each Aire de Sante which was the health facility catchment area

2. The Kinshasa province survey described in Burnett et al. had 3 sampling stages, what was the third sampling stage (hint: look under “Survey Objectives and Sample Size”?)

Ans:

The Third Stage Sampling (TSS) was households with eligible children for the selected age category

3. Based on the definitions of probability sampling and sampling frame found in the 2015 Vaccination Coverage Survey Reference Manual (section 3.6, section 6.2 and annex A), what do you think of the sampling frames used in country C vs. the sampling frame used in the survey described by Burnett et al.? Describe potential limitations of the frames used and how they may relate to sampling bias

Ans:

Sampling frame used in Country C

The sampling frame for country C planned to use the sampling frame for 2016 population and health survey. The frame contains the list of all Enumeration Areas (EAs) this frame will provide an opportunity to cover the entire population, and ensuring every household to be mutually exclusive. More so the enumeration areas have well defined boundaries, this will ensure the sample size to be sufficient to attain reliability and precision requirements.

Potential limitation and sampling bias

The likely limitation is that the household list was prepared by the field workers and may likely miss some areas especially hard to reach, isolated households or new settlements and may nit therefore be comprehensive. If the household list is prepared by these people there might be tendency of selection bias and even recall bias.

Sampling frame used in the survey described by Burnett et al.

The Sampling frame used in Kinshasa was a list of neighborhoods for each of the 12 zones with estimated target population which was actually driven from the polio microplan compiled from 2013 to 2014. Accordingly however this sampling frame do not have a well defined geographic or administrative boundary making it difficult to use the 15 neighborhood to inform household sampling. Random starting points were selected to correspond to the selected neighborhoods.

Potential limitation and sampling bias

In the case of Kinshasa sampling survey, the sampling frame do not have a clearly defined geographic boundaries, this migh result in missing isolated households. This may also result in high work load for some clusters. Which could result in selection bias and ascertainment bias thereby reducing the precision of the survey.

4. In the Kinshasa province survey described in Burnett et al., the expected sample size was not reached. The authors describe two potential factors that may have contributed to this. How could have this been prevented?

Ans:

The household list should have been more comprehensive to include all families regardless of whether they are single parents or not. In this case the DHS data should not have been used rather the zones should have been mapped out to have more accurate data. Better training of the surveyors could have been considered. Getting the telephone numbers of the caregivers might have been useful in arranging appointment with caregivers on the best time to find them at home. Conducting the survey outside major event such as festivals, rainy season or harvest season might be considered because during this type of period caregivers would probably not be home.

What are the main consequences of not reaching the expected sample size?

Ans:

If the expected sample size cannot be reached, then the result will not be precise and the findings will not be accurate and not necessarily valid for informed decisions. The objectives of the survey will not be reached, resulting in waste of time, energy and resources in the long run.