Digital Frontiers: Gender Data, The Private Sector, And User Consent
Earlier this week, I was a panellist at the Digital Frontiers event in London, hosted by DAI Europe. The event focused on digital inclusion, and how we can integrate vulnerable and marginalised populations into the global digital ecosystem and provide them with the information and service, and opportunities that access to digital technology can bring.
The event was a combination of lightning talks from the panellists - Krista Baptista from DAI, Guillaume Touchard from GSMA, Saloni Korlimarla from Cherie Blair Foundation for Women, and me - followed by small breakout groups to dive deeper into some of the topics raised, and then an open group discussion.
I’d never claim to be a big fan of public speaking - I’m much more of a one-on-one type person - but I particularly enjoyed this panel, because I got to wax lyrical about one of my favourite topics: the need for more gender-disaggregated ICT data and evidence. I wrote a few weeks ago about the launch of the Gender and ICT Survey Toolkit that I co-authored with Katie Highet and Hannah Skelley from FHI360, and why the first step towards closing the gender digital divide is to understand it - to quote Katie Highet, “If we don’t understand the shape and size of the digital gender gap, how can we effectively drive change?” And so we need more gender disaggregated data on ICT access and use - of which there is currently very little out there.
In our break out groups, we got to discuss this idea further, and share opinions, experiences and insights. Some of the key discussion points that came up was the role of the private sector in collaborating and sharing data - specifically, the tech providers such as Google, social media platforms like Facebook, and the mobile operators. In my experience, mobile operators do have very large data sets and insights, but they’re not always willing to share data, particularly as it is mostly confidential, and in my experience, very little of MNO subscriber data is gender disaggregated. Even if it is, it’s often not reliable - SIM registration data, for example, often masks the fact that men may register SIMs on behalf of women, so even though the SIM may be registered in a man’s name, it’s a woman using the phone, but that nuance just wouldn’t appear in the official data set.
Linked to this, the group discussed the role of social media in gathering gender-disaggregated data. Social media platforms may very likely have gender data - not just from their profile registration data, but also from data mining and profiling users based on their user activity, preferences and ‘likes’. The issue here is - what does this mean for confidentiality of user data and privacy concerns? This is a growing concern that is generating a lot of headlines - not just for gender data, but for all users. A recent New York Times article by a former Facebook privacy developer pointed out that Facebook “has no incentive to police the collection or use of data...Facebook is free to do almost whatever it wants with your personal information, and has no reason to put safeguards in place.” In our quest for gender-disaggregated data, what privacy or consent issues haven’t we considered?
At the other end of the spectrum, however, is a digital development solution that I recently created an M&E framework and methodology for. The solution doesn’t collect, store or track any individual user data at all, to ensure anonymity and confidentiality. While this is understandable and applaudable from a privacy standpoint, from an M&E, evidence and gender data standpoint, it means that we can’t understand our user base, and we particularly can’t understand our female user base. This means we can’t get any insights into the female user experience, where their pain points are, where their dropoff points are, and how we can improve their experience to ensure that they have equal opportunities to positive benefits access to this digital solution might bring. If we’re not collecting gender-disaggregated data, how can we ensure through our M&E that we’re reaching and engaging women?
This discussion around gender disaggregated data certainly generated a lot of thought - and I don’t necessarily have the answers. I do however believe that more gender disaggregated data is crucial in closing the gender digital divide (and I know I’m not alone here), but the event this week did generate a lot of questions for me. And one of them is, as ICT4D specialists and as proponents of more gender disaggregated data and evidence, are we having the right conversations about privacy? Are we considering ethics of collecting more data, at the potential expense of the privacy of the users? Throwing this out there to the readers - what are your thoughts?
