A decade of working with data in international development has taught me three key lessons:
Data without a strategy is dead. It becomes meaningless globs of numbers and stories easily blown about in a westerly wind. Data collection is very in-vogue right now. According to IBM, about 2.5 quintillion bytes of data are created every day, we love measuring everything these days, even how many times we blink in one day. Unless there is a foundation of strategy that guides the collection, analysis and application of the data it becomes an unnecessary burden on partners, and employees alike.
Data is not objective. There is a term in data science called veracity which measures the biases, noise, abnormality, and reliability in data-sets. John Giannandrea, Google’s AI Chief said before a recent conference that: “The real safety question, if you want to call it that, is that if we give these systems biased data, they will be biased.” Within international development, we need to identify the bias, communicate the bias, and prevent the bias through on-going feedback loops and asking key questions of how our data processes may be perpetuating bias.
Data needs to be shared. The international development community needs a large data bank on complex issues such as human trafficking, global hunger, and poverty. The data bank would have the capacity to hold multiple different data types along with metadata that describes the datasets. Data is often siloed away, becoming outdated and dusty.