Mapping the home locations of beneficiaries attending a program site is one of a number of simple ways of defining the actual (rather than intended) catchment area of each program site. The actual catchment area as shown by the map represents the extent of the spatial coverage of a program and indicates whether there is possible spatial unevenness in the program’s coverage.
The home location of beneficiaries is usually recorded on the outpatient treatment card but not routinely recorded in the routine data monitoring database. This should therefore be collected for use in mapping.
Mapping does not require the use of sophisticated mapping or geographical information system (GIS) software packages or the use of Global Positioning System (GPS) receivers. All that is needed is a paper map of the program area of useful scale with all or nearly all of the villages within the area shown.
Analysis of data
There are many ways of presenting the map of home locations of beneficiaries. The most easy and straightforward way is to tally the number of admissions coming from each village and then indicating this number on the paper map where the village is located. A visual alternative to this is to create circles on the village locations the size of which are proportional to the number of admissions from each location. An example of this is shown in Figure 1.
It would also be useful to map the home locations of beneficiaries who have defaulted as shown in Figure 2.
From FANTA technical reference page 28
From FANTA technical reference page 31
Another approach to presenting the home locations of beneficiaries on a map is through
choropleth map. A
choropleth map shades or colours or adds patterns to areas on a map proportional the value of the indicator being presented. This is only useful when the maps being used define borders or boundaries of the villages rather than just points (as with Figure 1 and Figure 2) or for urban area programs where maps would show city blocks. Figure 3 and Figure 4 are examples of
choropleth maps to show home locations of those who have been admitted and those who have defaulted (respectively) from a program in a district of Nepal with the village development committee (VDC) borders shown.
So as to preserve the paper map and to be able to use the same map to show different spatial indicators, transparent plastic sheets can be used as ‘map overlays’ to which the presentation of the data can be written or drawn. This approach not only preserves the paper map and making it usable for other purposes later on but also allows for multi-indicator analysis through multiple transparent sheet overlays. Figure 5 shows an example of this approach.
Photograph courtesy of Save the Children
To include a copy of the map onto a written report, there are two options available. First is to take a snapshot of the paper map with one or more of the transparent plastic sheet overlays (depending on the information that is being presented or reported) and use insert this into the report. The alternative is to re-draw the paper map and the various overlays on a computer using a vector graphics drawing software. Some coverage assessment technicians have reported using Microsoft® Visio and Adobe® Illustrator for this purpose. The downside of these drawing software is that they are expensive commercial products.
There are open source options, however. Figure 1 and Figure 2 are sample maps drawn using OpenOffice Draw OpenOffice suite) which is one of many open source solutions for vector graphics. A tutorial on the basic functionalities and usage of OpenOffice Draw can be found here.
Spatial coverage is an indicator of the program’s ability to reach out across the area it has intended to provide services to. An even or nearly-even spatial coverage is a sign of a program with high coverage. The program described in Figure 1 and Figure 2 has limited spatial coverage with coverage restricted to areas close to program sites or along the major roads leading to program sites (i.e., higher admissions and lower defaulting close to program sites while lower admissions, higher defaulting farther from program sites).
The program described in Figure 3 shows a similar pattern of higher admissions close to program sites. However, the spatial pattern of defaulters as shown in Figure 4 is not as straightforward to interpret. Mapping the absolute number of defaulters can be potentially deceptive. For example, comparing Belhi VDC (southwest tip of Saptari district) and Kalyanpur VDC (on the northern border of Saptari district) it can be noted that both have 4 defaulters and is represented with the same same colour intensity. However, Belhi VDC has only 8 admitted cases while Kalyanpur VDC has 42. In this context, these 2 VDCs are opposites of each other with regard to coverage even though they have the same number of defaulters. Kalyanpur VDC, which houses a program site, has a relatively better level of coverage based on a lower proportion of defaults with admissions (10%) compared to Belhi VDC (50%). Figure 6 maps the proportion of defaults with admissions and this shows more clearly the impact of defaulting with VDCs farther from program sites or without direct access to roads having higher rates of defaulting as a proportion of admissions.