How Big Data Is Helping Indian Schoolchildren
We often hear how big data and analytics are helping companies make more money. But big data tech is much more than that, and can also have a positive impact on one of society’s most vulnerable groups: rural schoolchildren.
Earlier this week, Hewlett-Packard shared a story about how it’s working with a non-governmental entity (NGO) called the Akshara Foundation to help improve the educational experience of schoolchildren in the Indian state of Karnataka.
Akshara’s mission is to ensure that every child has access to school. As Akshara foundation chairman Ashok Kamath explains, parents have a lot of choices for schools in big cities like Bangalore, which has a number of private schools.
“But the minute you go into the rural hinterland, the government school is often the only school,” Kamath says in a YouTube video. “And if we don’t improve the quality of teaching and learning in those schools, then we’re going to deprive an entire generation … of the benefits of a good education.”
As it so happened, the Akshara Foundation had a wealth of data to mine. Since 2006, the foundation has been collecting primary data about resources, facilities, and records from 40,000 schools in the state, with the goal of improving the math and reading skills of students at those schools.
“We believe in the maxim ‘What gets measured gets done,'” Kamath says. “A lot of decisions can be made by understanding data….Of course at Akshara, we don’t have all the skills, which is why we reached out to HP.”
Hewlett-Packard provided a group of skilled data analysts who could help identify the real-world triggers that lead to better learning outcomes for schoolchildren.
“We needed a way to turn this data into information that would enable us to make better decisions on which initiatives were helping students to learn and which weren’t,” Kamath says.
Over the course of nine months, a team of volunteers led by an experienced data scientist from HP’s Analytics and Data Management Practice named Rajesh Krishnan cleansed and prepped four separate data sources for analysis. Despite the size of the data and the fact that some of the data was messy or incomplete, the team found ways around it.
The data science team applied cluster analysis to see what patterns it could tease out of the data, at an individual classroom level. And patterns did emerge. For example, the data suggested that, to achieve success in the classroom, that schools aim for an optimal pupil-to-teacher ratio. Similarly, there is ideal number of books per child that correlates with success.
But data scientists also pulled some unexpected findings from the data. HP’s Krishnan explains:
“A valuable insight the dashboard provided was that engagement was particularly low among 11 to 14-year old girls,” Krishnan said. “However on further analysis we found that the introduction of separate bathrooms helped to significantly reduce drop-out rates. To have identified a way to keep this demographic in school is incredibly valuable.”
The program has been a big success for the Indian state of Karnataka. As a result of the project the HP volunteers were awarded the “2014 Global Volunteer Challenge: Most Impactful Program,” and also secured a grand worth $20,000 (USD) for the Akshara Foundation.
Now, the group is looking to see how the lessons can be applied to other areas.
“This has been extremely satisfying at many levels to see how we can use our data skills to help the overall impact on the society,” Krishnan said. “We plan to make this a program that can be taken across the country and try and see how we can ensure that every child in India gets a better education in the years to come.”