We all like to feel that what we’re doing is significant. As a nonprofit, doing something significant is probably one of your primary goals. You want to be making a difference in the world — that’s what matters.
But could you be making a bigger difference?
Is your organization able to accurately measure the impact of your activities and demonstrate the results of what you do? The answer may well be no — even if you think it’s yes. That’s not due to any particular failing on your part, but rather because data analysis for nonprofits tends to be a rather siloed affair. Information gathered is confined to analysis within the project it came from — and this isolation can cause larger patterns and trends to be overlooked.
You collect data on your individual project. At any given time, your team (or another organization with similar goals to your own) is running similar projects — and collecting similar data. How often do you combine data from projects that have common elements or goals?
Let’s examine three ways combining cross-project data can benefit you:
Increase Statistical Power
Making a difference in the lives of the people you work with is what being part of a nonprofit team is all about. But your resources are limited, and using them to greatest effect is critical.
What if you discovered there was only a 30% chance your work in a given area was having any effect at all on that community? Would you continue to direct resources to that project, or would you divert them to another project where you had a greater chance of making an impact?
Statistical power — the likelihood your data analysis will detect an effect if there is one to be detected — increases with the size of the effect and the size of the sample. While you have limited control over the size of the effect you’re trying to measure, combining data from multiple related projects can dramatically increase the size of your sample — improving your chances of accurately detecting changes brought about by your organization’s work.
Analyze Impact Without an RCT
It’s impossible to perfectly measure the effects of any project you undertake. Human beings cannot be relied on to perform consistently or predictably under the best circumstances. Differences in culture, politics, economics, and geography can all skew your data and leave you struggling to quantify your achievements — and wondering where to dedicate resources in the future.
Randomized control trials (RCTs) are an ideal solution to this issue if you’re a scientist in a lab whose primary goal is the collection of data. Data analysis for nonprofits, on the other hand, is a secondary goal — withholding your assistance to a portion of the population you’re working with is often out of the question if your primary purpose is to improve their lives. Running an RCT not only costs you valuable resources; it often reduces the number of people you’re able to help.
Combining data from multiple similar or related projects not only provides more data but also provides data for a larger sampling of people — meaning biases or effects specific to any one group can be more readily identified and accounted for in your analysis, without an RCT.
See Complexities Not Revealed by Siloed Data
There’s always more to the story than meets the eye. No matter how careful you are to keep your measurements precise and your data clean, the fact is that statistics gathered from a single project will never tell a story as complete as one using cross-project data.
There is a well-known story of the blind men and the elephant in which a group of men, all unable to see, attempt to examine an elephant using their other senses. While one ran his fingers over the inflexible lustre of a tusk, another manipulated the muscular trunk. A third traced his fingers over the soft down on the elephant’s ear, while the fourth focused his attention on the rough, callused skin of a foot. All four made completely accurate observations, but none correctly arrived at a complete picture of the animal he examined.
By combining data from multiple sources, you can shed light on data points across a much larger area, providing context for the information you’ve gathered and the conclusions you’ve drawn — effectively painting a more accurate picture. When it comes to data analysis for nonprofits, more data is almost always better.
In Action: CARE Cross-Project Analysis
This past summer, our team at Datassist had the privilege of working with CARE, an amazing humanitarian organization working to eradicate poverty and promote equality around the globe. They came to us for assistance in aggregating and analyzing data from three disparate projects, each with a distinct local goal but also part of a greater global goal: the empowerment of women:
Strengthening the Dairy Value Chain (Bangladesh)
Pathways to Secure Livelihood (Bangladesh, Ghana, India, Mali, Malawi, Tanzania)
Link Up (Kenya, Tanzania)
Siloed within separate projects, it was difficult to effectively measure the effects of the programs on the global goal; cultural and geographical differences hampered CARE’s ability to interpret the data they collected.
By helping them combine the datasets and looking at results across projects in seven countries, we allowed them to look beyond the context of individual communities and accurately identify subtle changes occurring across all groups. Using this information, CARE was able to further their goal of addressing gender balance issues in their collectives to better empower women around the globe.
Datassist: Data Analysis for Nonprofits
At Datassist, we know you have important stories to tell, and we understand that, as nonprofits, you don’t always have the resources to tell them as well as you’d like. Let us help you show your audience the big picture — so you can make a bigger impact.
Our team of statisticians, data visualizers, and graphic designers will assist you in collecting and analyzing data and translating it into beautiful visuals and stories that will engage, educate and inspire. Get in touch with us today to discuss how we can help you.