Identifying and selecting eligible individuals in the household
Rules for selecting individuals who meet different eligibility criteria need to be defined in the survey protocol. Methods include selecting all eligible individuals in each selected household, or randomly selecting one or more of the eligible individuals in a selected household, or selecting all or a random selection of individuals who meet certain eligibility requirements in every nth household, for example every 2nd or 3rd household. The agreed approach must be systematic and applied in the same way in all clusters. For example, if non-pregnant women of reproductive age is a population group of interest and there are three women meeting this criterion in the household, all three could be requested to participate. Alternatively, one of the three women may be randomly selected and requested to participate. (Note that in this case there would not be substitution by another woman if the first one refuses consent.) This methodology may be applied in all selected households or in a pre-determined number of households that are systematically selected from the household listing.
The decision on selection methodology is usually based on the target sample size for a specific population group, the proportion of this group in the population and the total number of selected households per cluster, as described in Module 5: Sample size. For example, in the 2015–2016 Malawi Micronutrient Survey 1where 22 households were selected per cluster, survey participants from different eligibility groups were enrolled according to the following schematic (see Fig. 7.4):
- All preschool children from all households.
- All non-pregnant WRA age from each of nine households.
- All school-age children from each of six households.
- All men from each of four households.
The required number of households for each eligible group was selected by systematic sampling from the 22 selected households. This schematic enabled the survey to meet the sample sizes required to achieve the survey objectives for each population group of interest, while ensuring that there was no unnecessary burden of purchasing and transporting survey supplies.
Fig. 7.4. Malawi 2015–16 data collection schematic indicating demographic group eligibility per cluster
Household 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 Preschool-age children X X X X X X X X X X X X X X X X X X X X X X School-age children X X X X X X Women of reproductive age X X X X X X X X X Men X X X X
National Statistical Office, Community Health Sciences Unit [Malawi], US Centers for Disease Control and Prevention (CDC), Emory University. Malawi Micronutrient Survey 2015-16. Atlanta: US Centers for Disease Control and Prevention; 2017 (https://dhsprogram.com/pubs/pdf/FR319/FR319.m.final.pdf, accessed 17 June 2020). ↩