Experience with CMAM programs shows that the distance or time that carers are willing or able to walk to access services varies greatly between settings. The half-distance between markets is a tool that is used to proxy this distance. It is estimated by first identifying hamlets, villages, and towns on a map. Then measure the distances () between the neighbouring villages and towns with markets and calculate the
mean (average) of these distances:
: Sum of the distances between neighbouring villages and towns with markets
: The number of distances between neighbouring villages and towns with markets
The distance that carers are willing or able to walk to access services will be approximately half of this mean distance.
A map of useful scale with all hamlets, villages and towns will be needed to perform the calculations described above.
Analysis of data
A worked example of the half-distance between markets approach is described in Figure 1.
Add the distances () together:
Divide the result by the number of distances ():
Divide the result by 2:
It should be noted that only distances between towns and villages with markets are used in this calculation.
The half-distance between markets approach should be used to provide a first estimate only. This estimate should be confirmed by other means (e.g., time-to-travel plots, discussion with carers and program staff). It is very important that the cultural and security context are taken into consideration. For example:
In some settings, women may not engage in trade or may not engage in trade outside of their home community. This often means that women are reluctant to travel far from their home community in order to access CMAM services.
In other settings, women must be accompanied by a male family member when they leave their immediate neighbourhood.
In other settings, it may be dangerous for women to leave their home community.
The half-distance between markets approach may overestimate the distance or time that carers are willing or able to walk to access services in such settings. The estimate should, therefore, always be confirmed by other sources and methods.