 Welcome to the ITU studio in Geneva. I'm very pleased to be joined in the studio today by Mr Ingmar Weber who is the research director for social computing for the Qatar Computing Research Institute for the Hamad bin Khalifa University in Doha. Welcome to the studio. Thank you very much. And now we're here at the WTIS-18, the World Telecommunication ICT Indicator Symposium and I wanted to start off by talking about numbers, a lot of numbers being banded about here this morning and I'm sure as the meeting goes on. But why is it necessary to keep track of numbers in the ICT sector after all many countries are now reaching saturation levels? Yeah, this is a good question and certainly there's a lot of emphasis on looking at things such as internet penetration, which is a very important indicator, but in our own work we try to look at a lot of disaggregated numbers. So for example, if you have a country with 80% internet penetration it could theoretically mean that you have 100% of men but only 60% of women with internet access. So just disaggregating these numbers is an important issue which is not always done. Another aspect that goes just beyond just counting looks at more qualitative issues like how for example the internet is being used, is it being used just for entertainment, is it being used for information, is it actually being used for economically useful activities sort of. So that's something that's also harder to measure and where maybe new methodologies could potentially help. In terms of non-traditional data sources, perhaps we could talk a little bit about those. How could they be used to monitor global ICT usage and in fact what are they? That's also a good question. So in our own work we use a lot of data from social media. In particular one data source we tap into are so-called advertising audience estimates. So for example platforms such as Facebook but also LinkedIn and for Contactee and any social media platform they provide advertisers with information on how many of their users match certain targeting criteria. So you can go to Facebook's advertising platform and you can ask, hey I would like to show an advertising campaign to women in Geneva who use the iPhone X on a 4G connection. Then before you launch your advertising campaign Facebook would provide you with information on how many people match this criteria. Now if you do this globally you can get a good sense of who's using this platform. For example how many men or women are using as a Facebook or LinkedIn in India or in other sort of countries. And what we observe in our own work is that looking at the in particular the gender ratios on these platforms is actually a very good predictor for modeling the gender ratios on the internet in general. So put very simply you know if you look at a country where you have far more men than women on social media then typically there's also a country where you have far more men than women on the internet in general. So here in Switzerland unfortunately country where there's gender parity but in many other countries of the world that's that's not the case. And what about the challenges of using these sources for data? Yeah so when you tap into social media in particular like one of the biggest challenges relates to selection biases. I mean of course not everybody's on social media I mean I'm myself I'm not on Facebook for example and so then it's a question well you know are you studying really society at large or are you just studying one particular platform. Now in our own work we sort of we we build statistical models to correct for these biases to ultimately hopefully enable us to reason not just about users of a particular platform but about all the I mean the whole population in a country. And what's important to note there is that even the fact that or in particular the fact that who's not on these platforms becomes part of the signal sort of as I mentioned before if we do observe that women are not on these platforms then that is the signal even though they are sort of ironically not part of the dataset. Now what about the fact that over half the world's population are now connected to the internet it was all over the news on Friday I just wanted to ask your opinion on that I think it's 51 percent is the figure that's being quoted but that's aggregated over the globe what's your opinion on that? Yeah I mean that's obviously exciting news and it's great to see but you know just as with let's say you know income going up in some countries you have to look at the distribution of that sort of a sort of you know who's benefiting who is online who's not yet online right so until you know it's 100 percent you know there's still 50 percent who's not on the internet and of course they're not equally distributed across you know geographies and genders and other dimensions. Nima, thank you very much indeed. Thank you for the opportunity.