Effectively Presenting Food Security Data
Are you also struggling with mapping results of food and nutrition security programs? In this blog I will show you possibilities to visualize your data. Not only because it’s a lust for the eye, but it can also help you detect interesting patterns.
Of course you can use regular tables to present your data, however if you use visualized data it’s easier to detect patterns. There are useful and easy accessible tools out there.
But before even thinking of visualizing you really need to clean up your data first. As I mentioned in my previous blog, it’s necessary to identify SMART indicators and gather a set of representative data. Here is ahandy link that helps setting your records straight.
To determine the food status of households, we use HFIAS survey. This survey allows you to categorize households into four categories, from food secure to severely food insecure over a period of time.
This is how we have gathered rich data sets, over the years. Instead of presenting these collected data in tables, ICCO programme officers and Monitoring and Evaluation (M&E) staff have started to experiment with using tools like Google Fusion Tables  and CartoDB . These tools make it easier to detect interesting patterns, which might be missed when regular tables are used to present the data.
When interesting patterns are visible, you will be inclined to search for what has causal relationships. You will also have to critically reflect on the relevance and effectiveness of your interventions. For example, the data might be used to compare districts where your organization is active.
The print screen below presents data for districts in India, wherebygreen= food secure, yellow=mildly food insecure, orange= moderately food insecure, and red=severely food insecure.
From the map, it becomes clear that certain districts perform significantly better than others. The identification of this pattern facilitates the further analysis of the monitoring data. It raises questions that the M&E officer together with the programme staff subsequently reflect upon, like:
- Was the baseline situation for different households and/or different regions the same?
- Why is our intervention working in one region, while it does not work in another? What are the underlying causes?
- Have there been external elements that have an impact on the results of our work?
- Have we selected the right target groups?
- Was our sample correct?
- What follow up measures are we going to take? Do we intensify our interventions in the districts that seem to perform relatively bad?
The further analysis of the significance and (potential) causes of the identified patterns can subsequently lead to an adaption of your interventions based on the insights gained.
Mind you: it’s not all fun and games
There are clear benefits of data visualization, like mapping, but there are some serious challenges. ICCO Cooperation has, for example, experienced that the analytical capacity at partner level might be (too) limited to critically reflect on the results produced.
Furthermore, as with other visualization tools, a risk of creating maps can be that users jump too quickly to (the wrong) conclusions based on the identified patterns, due to a limited understanding of the actual meaning and causes of the patterns.
Therefore, a key lesson learned for ICCO is that tools like Google Fusion Tables and CartoDB can be very useful as an entry-point for critical reflection and learning, but need to be accompanied by further analysis and research of monitoring data to effectively inform programming.