Why data analytics is central to education reforms
Harsh Garg
November 04, 2019

According to the latest Annual Status of Education Report, between 2005 and 2018, the proportion of Grade V students who could read a Grade II text declined from 60% to 44%. During the same period, the proportion of Grade V students who could do division declined from 42% to 23%. In simple terms this means that less than half the students in Grade V across the country are able to read and do arithmetic appropriate to their grade-level.

The implication of these stark numbers is that while India has largely been successful in solving the access challenge in education, the quality of education leaves much to be desired. Additionally, it means if we are to replicate the success of initiatives aimed at increasing enrolment, it is imperative we redouble efforts to improve the learning levels of students.

The question is, how?

The scale and complexity of the learning challenge in India makes the use of data critical for undertaking informed and targeted reform efforts. State governments have to be judicious in deciding how to collect the data they need and to what end they will use it.

Granular and accurate data can change the way governments approach programme design, make operational decisions, and carry out review and monitoring. This is borne out by the experience of Haryana, Himachal Pradesh and Andhra Pradesh in the reforms they have undertaken to transform their government school systems.

Data-backed programme design can go a long way in ensuring the eventual success of any reform effort. Before rolling out a state-wide remedial programme, across more than 8,000 primary government schools in the state, the Haryana government commissioned an impact assessment study on a sample of around 3000 schools. During the study, it was noticed that performance in English, which at that point was not part of the remedial programme, was extremely poor. Feedback from teachers indicated that teaching different subjects each day, as was the prescribed practice at the time, meant that lessons taught were disconnected from each other. Third, some concepts mentioned in the training manuals for teachers did not have sufficient practice content to ensure students grasped specific concepts. Thus, before scaling up the remedial programme to all schools, the Government of Haryana made several changes to the programme’s design. They created teaching content for English remedial classes, prescribed teaching the same subject on two consecutive days, and added more worksheets and activities in training manuals, especially for the concepts flagged by teachers. This not only helped mitigate the programme’s shortcomings but also build consensus among government school teachers to effectively carry out remedial classes. Currently, the remedial programme, called Learning Enhancement Programme is being implemented across nearly 97% of all primary government schools in Haryana.

Besides programme design, data analytics is also crucial for operational decision-making across the education system. In Haryana, the Saksham Adhayapak Dashboard, which captures the academic performance of all 14,000 government schools on a bimonthly basis, is used by different stakeholders to make operational decisions. For instance, at the district level, Deputy Commissioners use this dashboard to identify the granular reasons why some classes in some schools aren’t doing well or why students are doing well in a particular subject as opposed to others. At the block level, given the number of schools that block education officers need to visit regularly, they use data from the dashboard to identify and prioritise low-performing schools. At the school level, school heads use data from the dashboard on an ongoing basis to discuss student performance with teachers, identify weak learning outcomes and prioritise topics to be retaught. The dashboard attracts nearly 2,00,000 views per exam result (released once in two months), thereby demonstrating not only the utility of capturing data on student performance but also for driving improvement.

Lastly, data enables effective review and monitoring, essential for tracking progress and taking timely corrective action. As per guidelines under Sarva Shiksha Abhiyaan, each state is supposed to establish a monitoring mechanism where the Block Elementary Education Officers (BEEOs) and Block Resource Centre Coordinators (BRCCs) visit a certain number of schools in a month and fill a Quality Monitoring Tool (QMT) focussing on issues related to administration, infrastructure and learning levels. In Himachal Pradesh, this information was captured offline and mostly not discussed at any state or district review meetings. This, in turn, led to a lack of accountability towards both the monitoring visit compliance and improving the academic performance of the state. In 2017, the Government of Himachal Pradesh introduced an OMR-based monitoring form to digitise this data, replaced by a mobile application in 2018. In addition to this, the government also made it mandatory for all districts to conduct a district meeting every month for reviewing monitoring compliance and the issues flagged during these visits. This has resulted in nearly 90% of all government schools in the state being monitored at least once in the past 12 months. Moreover, the state’s education department has been able to resolve 76% of over 6,000 identified issues flagged. This has been possible because data on school visits and the issues recorded are collected centrally and available for the Department of Education to check for non-compliance.

It is important to note here that data is not just meant for the use of officials at the state, district or block level. Teachers, the most critical change agents in the education system, can carry out their duties more effectively if armed with the right data. Andhra Pradesh’s education department has taken several constructive steps to this end. The state government wanted to ensure that teachers across the state used assessment data from the National Achievement Survey (NAS) and the Student Learning Achievement Survey (SLAS) to identify areas of improvement with respect to student performance. However, it was imperative that they were conveyed these insights in an easy-to -understand manner with actionable recommendations. Therefore, the state’s education department created short videos with relevant insights from the NAS and SLAS results and disseminated them to all the 346 mandals (districts) of the state using 836 virtual classrooms through a workshop. More than 13,000 teachers from nearly 30% of government schools in the state attended the workshop. The feedback for these videos, collected from more than 5,000 teachers, was exceedingly positive indicating the utility of equipping teachers with relevant data to improve learning levels.

The experience of Haryana, Himachal Pradesh and Andhra Pradesh shows that collecting and using data effectively can be key to implementing successful education transformation programmes. While data is not sufficient by itself, it is undoubtedly an enabler. As the saying goes, you can’t improve what you can’t measure.