O acesso à internet pelo mundo (Matéria em inglês)
Global map of active Internet subnetworks for the year 2012 - Map by Nils Weidmann, background from Natural Earth
Marginalized ethnic groups have poorer internet access even in the same country
This holds even when taking wealth and remoteness into account
Around the world, marginalized ethnic groups have worse internet access than more dominant groups in the country, researchers say. This is probably because of decisions made by government-run telecom providers.
Researchers mapped internet access and compared that to a list of ethnic settlement regions to show that less powerful ethnic groups have worse access, in a study published today inScience. It turns out that an excluded group usually receives about 60 percent of the average level of access across groups in the same country; its members are also more likely to have no internet access at all. That’s even after taking into account factors like wealth and whether these groups live in a remote area.
The internet is supposed to be a “liberation technology” that helps people organize for democracy; the most famous example is that people used Twitter to organize during Arab Spring. Still, there’s a “digital divide” in who has access to the internet and who doesn’t. Previous work has focused on how different countries or income affects access. Today’s research shows us that those in charge could subtly use internet access (or lack thereof) to keep themselves in power, and it serves as a reminder that technology has its own biases.
RESEARCHERS USED MANY DATA SETS TO MAP HOW LEVELS OF INTERNET ACCESS FOR ETHNIC GROUPS
“A lot of people still think unconditionally that these technologies can bring about peace, democracy, and prosperity, but we shouldn’t take for granted that everyone has access and the positive effects will just unfold,” says lead researcher Nils Weidmann, a professor of political science at Germany’s University of Konstanz.
The first step was figuring out which areas had good internet access. Comprehensive data is only available for countries — and the researchers wanted to look at ethnic divisions within a country. (Regional statistics are limited.) So they came up with a new method using computer IP addresses, which attaches to every computer connected to the internet. The harder problem is figuring out how many of these addresses are active.
Think of the computers like houses on a street. Because we have the addresses, we know that houses one to 20 exist, but it’s possible some houses are abandoned. To make the estimation more precise, the team analyzed all the internet traffic coming through a big Swiss network provider, which works like the post office for our metaphorical neighborhood. Post office workers can’t see the houses or the street, but they see which addresses are sending or receiving mail; these addresses are more likely to have someone living there.
Using this data, the scientists created a map that showed levels of internet access around the world. Then they compared this first map to a different one that showed where certain ethnic groups lived. This let them figure out which groups have better internet access.
The researchers used a list of politically relevant groups called the Ethnic Power Relations list to determine which groups were marginalized. Groups that have meaningful government representation are listed as “included” and those who don’t are “excluded.” The researchers also controlled for wealth, geography, and the region’s remoteness to make sure the differences they were seeing were because of political and ethnic bias.
This was an innovative way of figuring out how to work with the limited data on internet access within countries, says Andrea Calderaro, director of the Centre for Internet and Global Politics at Cardiff University. The results fall in line with what researchers suspected, but this is the first time there was quantitative proof.
INTERNET ACCESS HAS TO COME FROM SOMEWHERE, AND THAT “SOMEWHERE” CAN BE BIASED
There are plenty of other questions this finding brings up. One is that no one knows why we see this effect: it could be that the government “saves” the internet for its favored constituents, or that they’re deliberately keeping it away from ethnic groups that are more likely to rebel. Nor is it clear how this inequality is implemented. Maybe, says Weidmann, someone in the administration suggests that a telecom company focus on one region instead of another. But it’s very hard to find evidence because such decisions are so informal.
These results are a reminder that — for those trying to decrease the “digital divide” by working with government — there’s a risk the governments themselves are using aid to reinforce digital inequality. “It’s basically known that in several cases governments can use aid that comes to a particular country in order to channel it to those constituencies and populations they favor,” says Weidmann, “and that’s something that we don’t want and should be avoided.” The United Nations has declared that internet access is a human right, but we have to keep in mind that internet access has to come from somewhere, and that “somewhere” can often be biased.