Embracing Data-driven Approaches in Non-Profits:
In last month’s blog, Alan wrote about how non-profits can become more data-driven and base their services on the evidence. As GtD’s senior data analyst, I am often working with Alan and our non-profit clients in the US to develop tools and apps to help our clients become more “data driven”. For some clients, that means tailoring our ValiDATA tool that allows them to measure their services and outcomes. For others, it is developing dashboards to understand local problems and how they might respond to them better. This month, I want to highlight a dashboard we developed for one client and how we used APIs to open up a world of data from various open sources.
The Rich World of Open Data:
The world is full of exciting and freely available data, some of which could very well be useful for your organisation. You might want to use government statistics to better target interventions by demographic or location. Or perhaps an NGO has published open-source data that could help your organisation to predict outcomes better.
An API is a way of getting data from one place to another. API stands for “application programming interface”, and it allows different computer programs to speak to one another. If you imagine an external dataset as a reservoir, then APIs are the pipes that carry the information to where you can use it. By using APIs, you can save time and effort compared to manually downloading and processing data from websites, and you can also automate the process of retrieving and using data from external sources.
APIs: The Data Conduits:
One of the most exciting things APIs can do is collect data from multiple sources and collate it to produce new data. These can then be represented in a visually attractive application or dashboard to help understand local problems. To illustrate how this works, let’s look at a project that Get the Data did for a non-profit client who works to widen access to health care in rural Georgia.
A Case Study in Healthcare Accessibility:
When the client contact us, they asked us if we could provide a tool that would show how long people in rural Georgia would need to travel to access their nearest healthcare provider. After doing an exploratory data analysis, we devised a simple, easy-to-use solution. By combining data from the US Postal Service with a list of clinics and healthcare centres in Georgia, we created a tool that allows users to click on a ZIP code in an interactive map and then see the average time taken to drive to the nearest healthcare centre.
Using data from the latest US Census and the Robert Wood Johnson Foundation and overlaying them on a GIS (a geographic information system, a way of visualising data associated with a geographic location – think Google Maps), we added a function that allows users to click on any county in Georgia to see information about the area’s demographics and the healthcare needs of the population. This is a good example of how combining data from different domains (healthcare, demographics, and geography, in this instance) can yield new insights.
Creating User-Friendly Data Tools:
Tools such as the one I described are designed to help our clients identify challenges, allowing them to plan more effectively and target resources more efficiently. These types of tools can be readily created by taking data from external datasets, cleaning them up, and generating analyses. The real magic to “getting the data” is then to build a user-friendly web app so that users can access the information via any browser quickly, easily, and without time-consuming or expensive training. It gives me a great sense of satisfaction when I am able to give users access to the information they need when they need it.