Using data to investigate inequality, and building a network to find solutions
At a recent Open Data Day hackathon in Mongolia, a community grew around their exploration of place, pollution, and transit
On the map of Ulaanbaatar, the capital of Mongolia, the growing inequality isn’t that hard to notice. The ger district, a district of traditional yurts on the city’s outskirts, is home to many migrants who have moved to the city from the countryside. I mapped all the traditional yurts in my city a few months ago, and it hit me hard. Ulaanbaatar has clearly been divided in two: the poor and the not so poor; those with basic services such as water, heating and sanitation and those without; those who have access to bus stops and those who don’t, and so on.
The Data Lens datathon, organized to celebrate Open Data Day in Mongolia last month, tried to bring attention to inequality issues like these. Teams of data scientists, college students, journalists, and others competed in three categories, all highly relevant to the inequality challenges the coldest capital on earth faces today: air pollution, traffic, and public transportation. And the results were amazing.
Using data to investigate inequality—and explore solutions
One team analyzed recent indoor air pollution data from three kindergartens, two of which are located in the ger district where many children live in traditional yurts. They identified that children from the ger district kindergartens breathe in 1.5–1.8 times more polluted air than children in the city center, which does not have any yurts in its 1.4 km radius.
But even the downtown kindergarten children suffer from high indoor air pollution—about 50% above the national air quality standard. The team also determined that levels of carbon dioxide in the indoor air are at a hazardous level due to the large number of children in Ulaanbaatar kindergartens. “You open the window, and you get PM2.5 pollution from the outside. But if you don’t open the window, you also get high levels of CO2 pollution from children’s breath,” said Enkhmanlai, one member of Team Debuggers. “Opening a window in that situation is really like a double-edged sword.”
Prior to the finals, teams had the opportunity to present their ideas to eight mentors who are experts in their fields, and on the final day, they pitched their polished results to the jury of five open data experts. Some finalists came up with potential solutions to the problem they observed with their mentors.
Team Flame, the winner of the datathon, is a good example. Like the four other finalists, they picked public transportation. S.Dagvadorj used bus commute data to determine where passengers boarded and got off—according to his results, most commuters start their trip from the outside of Ulaanbaatar’s city center, while their final destinations are downtown. The ger district is home to 60% of the city population, so this makes sense. They have to travel downtown because the majority of schools and service centers in Ulaanbaatar are located there, forcing some buses to carry 2.9 times more people than their capacity in the mornings.
“One way to solve this is to have more buses. But we shouldn’t have more buses starting from a single point like we do now, but to run them simultaneously from the multiple busiest locations,” Dagvadorj said. The team also found that more than half of the households in Ulaanbaatar don’t have a bus stop within 500 meters of their home, pointing out that we need more bus stops in the outskirts of the city.
Another team of finalists used public transportation data to explore a completely different topic. Team Golden Insight identified students’ home and school locations by determining where they got off the bus. Not surprisingly, most students start their commute from the outside of the city center and end up downtown.
So the question is: Do these students not have schools nearby? Why are they going somewhere else for education? Using additional data from the Ministry of Education, team members found that some districts don’t have many school-age children, but still have overcrowded schools. But there are also many districts without enough students to fill their schools. The overcrowded districts include Ulaanbaatar’s best schools—known for national awards and higher scores on standardized tests. Parents can choose where to send their children to school, and the team concluded that one reason our public transportation is overloaded is because of these inequalities in educational resources.
Building a community
Overall, it was nice to see people from data journalists to programmers come together to make one data community. The judges announced data forums they are planning to organize and invited everyone to come. “This is Mongolia’s first datathon, and it exceeded my expectations,” said B.Batdavaa, a datathon judge and the director of the National Statistics Office. “I think organizing the Data Lens datathon every year will facilitate the growth of the data community in Mongolia.”
One can say that the Mongolian data community right now is small and quite fragmented. The truth is we often find people from different data companies collecting the same data without knowing about each other. But this event helped connect the community, and as the winners were announced, people handed each other their name cards, offering to work together next time.
Delgerzaya is a producer at TenGer TV, where she focuses on environmental and data journalism. She produces a data-based news segment, “Data Duran” at TenGer TV. With a background in climate economics and environmental research, she values the importance of data in revealing inequality issues in Mongolia.