What does it mean to be a Data Scientist in the humanitarian sector?
Back in 2012, Harvard Business Review stated that Data Scientists have the “sexiest” job of the 21st century. They also named the role of a Data Scientist as the second fastest growing job in the U.S. market. And it is true, Data Scientists are in high demand. People often view this role as a data solution master, when in fact, we usually come up with new problems and more questions than solutions. Nevertheless, being a Data Scientist in the humanitarian sector is indeed an exciting job. The most important part though is having the opportunity to use your skills, mindset, and tools for social good.
What is a Data Scientist?
There is no universal definition of a Data Scientist. The general agreement is that a Data Scientist is a sort of interpreter with a toolbox. Data Scientists have the ability to translate back and forth from technical jargon — usually related to math, statistics and/or computer science — to business strategy or sectoral expertise. And beyond translation, the interpretation: the ability to communicate the data insights found — visually or in other creative approaches.
A Data Scientist is a human (yes it is important to make this distinction nowadays) that can support others to solve problems or respond to critical questions by analysing and finding trends in data, both structured and unstructured data (usually referred as “big data”).
Examples of these types of data include:
- A single spreadsheet with a survey or Geographic Information System (GIS) coordinates (small-structured data);
- Text transcription of a focus group discussion (small-unstructured data);
- Sensor data with per second timestamps, or call centre logs (big-structured);
- Voice recordings, social media/media posts, and satellite imagery (big-unstructured).
In summary, a Data Scientist should be able to collect, clean, process, analyse, and visualise all of the aforementioned examples of data.
From data wrangling to changing mindsets
Shelley Palmer, a Data Science Adviser, created a Venn diagram on the minimum basic skills needed for a Data Scientist: computer programming, subject expertise, math and/or statistics. An important point missing here is the criticality of communications; the ability to visualise and turn into action some of the data research findings and ideally influence decision makers to turn these insights into action. We’ve updated her initial diagram to reflect this crucial competency.
But beyond the skills, a Data Scientist requires a particular mindset with multiple important facets. This includes:
- A mindset that emphasises detail upon analysis but the big picture on communication;
- A mindset that is inquisitive; the ability to dive deep into conversations with colleagues to obtain expertise that they have on their data;
- A mindset that values principles, to help others reform processes that are related to ethics, transparency, and accountability.
This mindset is important because even if you are “technically” savvy as a Data Scientist, nowadays a machine could process data faster than us. Still, the mindset of curiosity, putting ethical frameworks first, doing no harm with data, and the ability to communicate insights to push for social good, is the realm of humans in this field. This is actually the main idea behind the fourth industrial revolution: it all comes down to people and values. And in our sector, people and values are the highest desired competencies. The work we do reflects our values and we bring value to people with our work. This could translate into building a map or a graph to help scope the magnitude of a humanitarian crisis or by analysing social media text to provide insights into appalling xenophobia, discrimination and racism towards refugees. Our mindset defines our worth as Data Specialists. We can do our job better if we understand people and put people first.
The honest truth: our challenges
So, what are our challenges working in the humanitarian sector?
First, is the lack of full research freedom, compared to more academic fields of work. This is challenging for Data Scientists whose curiosity has driven their research success. In the humanitarian sector, we pursue research because there is a humanitarian need. And with that need comes the responsibility of delivering timely insights.
The second challenge is the complexity of the issues we research. Imagine trying to communicate the complexity of human behaviour — like the intention to flee for refugees — even in zones where there is clearly a conflict and the data clearly portrays that people are not moving. One phrase I keep repeating in the office is: I probably can’t tell you why but I just found a correlation although that might be not the cause.
Third, is that our sector is often lacking high-quality data. Paradoxically, this is the most important thing we need to do our work. There are many reasons why we don’t have the quality of data we need. Sometimes data access is a constraint because of individual privacy and protection principles. Other times, no one is collecting it because there is no humanitarian access to the area where the data lies. And sometimes the methodology for data collection is simply just poor.
Another critical challenge for Data Scientists is the need for more diversity in our sector. The technical expertise needed to become a Data Scientist usually comes with studies related to science, technology, engineering and math (STEM) areas — backgrounds that have been typically dominated by a male workforce. Data Scientists who are women encounter challenges that male counterparts don’t face. We struggle to have access to equal space to speak about our work and consistently face an overarching male-driven narrative.
For example, I recently participated in a panel on socially inclusive Artificial Intelligence (AI) at the AI For Good Summit in 2018. Despite the majority of the panels at this conference not representing the diversity of the sector, our panel had primarily women speaking about their experiences. Only a few minutes into the discussion, the panel was interrogated about why we were speaking about diversity and inclusion when the panel only had one male speaker versus four female speakers. The audience member was offended by the notion that four women and only one man could represent a coherent voice on diversity in the data science and AI space.
While I understand the sentiment, I disagree with this shallow view of equity and diversity. The majority of people designing these systems are white and male. When it comes to designing AI, we need more women and we need more diverse voices building these systems — otherwise, they will be inherently biased. This challenge is not only related to creating inclusive AI but a sector that values and rewards a diverse workforce to take on the opportunities of science, data and engineering in a complex world.
We are here to stay
Even with these challenges, Data Scientists in the humanitarian sphere are here to stay. We still have a long way to go before people truly understand how Data Scientists can add value to the humanitarian sector. We will also need to change the process of hiring and change our HR policies to be more flexible and to attract the talent needed, and more importantly, retain it. It also requires bold managers to bet for systemic change, to bring us on board and challenge what a traditional humanitarian looks like. For example, in my case, my manager bet on me, invested in me and gave me access to learning opportunities to bring new knowledge into the team and organisation.
And as a Data Scientist, we can’t create change alone once we’re inside a humanitarian organisation. I work with a great team of engineers to help build the data portfolio for UNHCR’s Innovation Service. Not every Data Scientist can have math, computer science, and engineering skills — seek people with skill sets that complement your work; collaboration is key.
So, be bold, be humble, and if you’re a manager — create space for non-traditional profiles in your team so you can find your unicorn.
This essay was originally posted in the recently released publication — UNHCR Innovation Service: “Orbit 2018–2019”. The publication is a collection of insights and inspiration, where we explore the most recent innovations in the humanitarian sector, and opportunities to discover the current reading of innovation that is shaping the future of how we respond to complex challenges. From building trust for artificial intelligence, to creating a culture for innovating bureaucratic institutions and using stories to explore the future of displacement — we offer a glance at the current state of innovation in the humanitarian sector. You can download the full publication here. And if you have a story about innovation you want to tell (the good, the bad, and everything in between) — email: innovation@unhcr.org.
