PERCEPTION SURVEY METHODOLOGY
To undertake the Community Perception Survey, 40 enumerators were trained over five days and deployed across the 14 priority earthquake affected districts to collect data over the course of 12 days from a total of 2100 respondents using a probability proportionate to size (PPS) methodology. All data collection is completed with mobile tablets using KoBoToolbox.
All VDCs in the 14 priority affected districts in which 60 percent or more of the households are eligible for the housing reconstruction grant will be considered part of the survey’s operating area, and eligible for random selection.
The population of each district will be considered the total population of all eligible VDCs, as per the 2011 census. The first 2000 samples of the survey will then be distributed by district proportionally.
The remaining 100 surveys will be allocated to districts where the total proportional sample size is under 100 respondents, in order to boost the population for an adequate district level analysis of the findings.
The number of VDCs selected in each district will vary, depending upon the number of samples allocated to each district. Each VDC will have a minimum of two wards sampled, and each ward a minimum of 10 surveys completed. Both VDCs and wards will be randomly selected from the list of eligible VDCs.
Twenty-five percent of the total sample will be allocated for municipalities, and municipalities will be randomly selected where there is more than one municipality in a district. In municipalities a minimum of three wards will be sampled, with a minimum of 10 surveys collected per ward.
Selection of households and respondents
Once wards have been selected, enumerators will identify an entry point in their given ward, targeting aschool, temple or other communal spot to initiate the individual interview process. At that point they will spin a bottle. The enumerator will walk in the direction the bottle points to once it has finished spinning until a home is found to initiate the interview process.
The first house selected will form a basis to identify other households to complete the survey of that ward. After identifying a first house for interview then enumerator will leave the house, turn right and skip the next two houses, completing the next interview at the third house. The enumerator will have leverage to move to next adjoining ward to complete the interview process if in the ward the sample household numbers are not covered.
Once in the household enumerators interviews an individual aged 15 years and above from the pool of all eligible respondents present in the home at the time of the survey. The enumerators select respondents from different age groups and genders at each home, to ensure the sample is demographically diverse and reflects the population from the survey area.