Anyone with experience conducting surveys in Kenya (or similar places) and needing to do revisits to specific households? Any particular methodology for finding a place again (both rural and urban) when addressing might be lacking? This question is less about longitudinal data collection (linking the subsequent surveys to one of the initial surveys), and more about collected-data and tools that can be used to find a household again if you don't have something like "123 Main Street, Central Town" to follow.
I worked with a research team working in informal settings without addresses..... I could find out and revert.
That would be great, please do!
My organization does a lot of this and I will ask as well, but so far as I know, what we have done was to both take GPS coordinates - obvious but not always too accurate, given the density of some informal settlements - but also ask a member of the household, "How would you tell someone how to get to your house from --insert known landmark--" and the combination met with pretty good success.
I'll ask more tomorrow from some of the researchers.
In addition to what Craig has stated, asking about the closest landmarks (usually schools, churches, health facilities or administrative offices) to there households together with their mobile phone numbers (and alternative phone numbers) works well. Initially, we used to ask them to draw a map to there households but that had its own challenges just like the GPS.
Also, depending on the type of project you're conducting, the local administration (e.g. Chiefs & Community health workers) can come in handy if they have the recruitment data to the households.
Hope this helps.
Becos there is address there is nothing for any geocoding algorith to use in doing reverse geocoding in locating those household. Ask for the said landmarks, stream, lake, market, townhall or palace. This may help. There go with GPS and get x, y for future or use mobile phone if there is telecoms there.
Hi! I recommend capturing geocoordinates and then using maps.me to navigate back to the household if you need to return. That app provides directions even when offline. More informally, I've done follow-up in rural areas by using lists of the entire household (all names) and as many phone numbers as possible. In rural areas I find that my enumerators can "ask around" using this information or call the numbers and then find the households. Honestly even when I have the address the above 2 methods have worked better, since no one really uses addresses in those areas. Hope this helps!
@DaniceBrown's recommendation reminds me of this related thread from a while ago that has some overlap: What is the best way to help direct data collectors to a destination?.
do you know if others gps app had been connected to ODK (GPS compass, Waypoint finder...) since 2019.
wa are actively looking to use GPS datum in odf form to launch a waypoint activity with gps app available.