Sarah Telford from UN OCHA on the challenge of open data in humanitarian response


In the humanitarian sector, we cannot be effective if we don’t know what’s happening. We need to know where people are, we need to know what they need, we need to know who’s there. So, data is essential for our work. My name is Sarah Telford, I work for OCHA and I’m the lead for the Centre for Humanitarian Data in The Hague. I used to work on sitreps and different information products and what I saw was that they were mostly all narrative, block narrative reports, and I thought, I wonder if we could just get a little more analytical with some of these reports. And then I went and I tried to find the data and I couldn’t find it. It was really all over the place. You know, someone would have it in a local file or in their laptop, and I just thought, you know, this has to be easier we have to be able to make it easier to find and use data and so that’s how we created the Humanitarian Data Exchange, HDX. So, we created that in 2014, we now have 6,000 datasets from 350 organisations, and you know, now, if I want to know what’s going on in Afghanistan I can go there, I can look, I can see who’s doing what, where. I can find the boundary data. So, it’s helping. I think where we need to go now is trying to bring all that together for inside, and that’s a lot harder because of the fragmentation, because there’s so many players. The thing that keeps me going obviously is that we cannot be effective if we don’t know what’s happening. We need to know where people are, we need to know what they need, we need to know who’s there, who’s responding, what are the gaps, so, data is essential for our work. At HDX, we’ve never allowed personal data to be shared through the platform. And so, we have a quality assurance process where you can check. We open a file, we see if there’s any names or any personal information and we can see that really obviously. I think where it gets more complicated is when you’re dealing with what we call community-identifiable information or demographically-identifiable information and the problem there is it’s not obvious when you open the file that there’s any risk of re-identification. For instance, we recently had some data come in from a certain camp where there was a survey done. And you’re seeing more of these community-perception surveys and we’re really happy that that’s happening, but when we looked at the file, we didn’t see anything wrong with it. Then, we got an alert from a colleague of ours that said, you know, if you do analysis on this, it’s a really high risk of re-identification and you could actually locate women who’ve been sexually abused and where they’re located in camps. And I was horrified. I just thought, oh my God, this is the reality now that it’s not apparent that this data is dangerous, but it is, or it could be in the wrong hands. So, this is why now we’re turning to this kind of idea of HDX Secure, which if HDX itself was challenging, which it really has been of just getting people to share our data and, you know, come on the platform. HDX Secure will be a hundred times harder because any time you’re trying to exchange data in a secure way, it’s really users, you know, with the passwords and encryption. And how this is handled, I think it’s really gonna be challenging but I have, I think, we have to go there, I think we have to start to set up a secure infrastructure for the exchange of data in the humanitarian sector. So, just last week, we were in Amman, looking at the whole of Syria operation and how data’s being shared from the different hubs in Damascus to Gaziantep to Jordan and you know, they’re doing an amazing job really, they really have clear protocols about the level of data, community-level data is only allowed to be shared at the hub level. You know, I was really impressed, but I also saw that that fear and sort of confusion around what data is safe and what data isn’t safe makes it so that no data is shared, you know, publicly. And so you get the situation where donors might go and say, you know, we really, we’d like to see this data, it’s in your product, can we have the file, and this just sets up this conversation, and you think, gosh, this has to be easier. You know, it’s a huge, huge risk and certainly in Syria, I mean, what we saw or what we heard was, you know, not only is it about the data, about the affected person and that kind of personal level data and household level data, but also the people who are collecting are really at risk. When there’s a change in control from government to non-government and you have people that have been collecting data or supporting, you know, hospitals and health care, then then they’re at risk for having supported or seen to be supporting or having information that could be valuable. So, this is really the crux of the problem, right, in these war zones, you know, and the duty of care that exists, the ethical dilemmas about how important it is really to collect the data? I mean, we know, Syria’s terrible, do we really need another needs assessment if it’s gonna put someone’s life at risk? So, these are the questions, I think, we have to navigate.

Leave a Reply

Your email address will not be published. Required fields are marked *