• September 10, 2024

Closing the Learning Gap: Strengthening FLN Through Ei Mindspark’s Personalised Learning

Mindspark is a Personalised Adaptive Learning (PAL) platform developed by Educational Initiatives (Ei) in 2008 to address learning gaps among students for Grades 1-10. Ei Mindspark is one of the partners on the LiftEd EdTech Accelerator portfolio which is supported by the Michael and Susan Dell Foundation, Reliance Foundation, UBS Optimus Foundation and USAID as Founding Partners, the British Asian Trust as the Programme Leader and Central Square Foundation (CSF) as the Design and Technical Partner. The LiftEd EdTech Accelerator aims to significantly shape the future of tech-based learning at home for foundational literacy and numeracy (FLN) in India. It will catalyse the supply of pedagogically sound and contextually relevant solutions while galvanising demand for at-home learning. This blog highlights how Ei aims to improve the efficacy and engagement of Mindspark for children in early grades by strengthening the Foundational Literacy and Numeracy (FLN) content on the platform.

India has one of the largest education systems in the world with over 26.5 crore students. Over the past two decades, efforts by a number of government programmes have led to a near-universal enrolment rate in primary schools[1]. While the progress around access to education has been promising, challenges around the learning levels of students remain[2].

In India, ASER 2022 data reports suggest that 80% of children in Grade 3 cannot read a simple Grade 2 text with meaning and understanding, indicative of the scale of the learning crisis. A growing body of evidence[3] shows that computer-assisted personalised learning solutions may be the most promising approach to improve student learning outcomes, especially among those who lag behind expected grade-level competencies. At the forefront of such PAL EdTech tools in India is Ei Mindspark. 

Ei’s Journey with Mindspark 

Ei developed the Mindspark app in 2008 using numerous insights collected over the years from student learning levels and learning misconceptions[4] among students gathered through large-scale assessments. Leveraging this data, Mindspark creates individualised learning paths for its users, providing targeted instruction to enable ‘Teaching at the Right Level (TaRL)’ at scale[5].

Contextualised literacy and numeracy content on Mindspark

In 2012, Ei set up Mindspark learning centres for low-income communities in South Delhi, with CSF’s support, which also partly funded the development of Mindspark Math and language products in Hindi. From 2015-2016, a randomised controlled trial was conducted by the researchers of Abdul Latif Jameel Poverty Action Lab (J-PAL), Massachusetts Institute of Technology led by Professor Karthik Muralidharan, (Co-chairperson of Education Research) to test the impact of Mindspark on student test scores in Math and Hindi. The study was conducted among 619 students between grades 6 and from five public middle schools close to the Mindspark centres. The results showed significant and promising improvements of 0.37 standard deviations in Math, improving by over twice as much as students in the comparison group and 0.23 standard deviations in Hindi, improving by 2.4 times as much as students in the comparison group over just a 4.5 month period. The relative impact of the programme was found to be much greater for low-achieving students, who were making no progress in school. The findings stated that these were among the largest gains of any education intervention that were observed in India.

Students at Mindspark centres in Delhi

A subsequent RCT was also done by Professor Karthik Muralidharan and Prof. Abhijeet Singh (Associate Professor of Economics, Stockholm School of Economics) to evaluate the impact of the programme on classroom processes and student learning in Adarsh schools, in four districts of Rajasthan, over three academic years. Between the start of students using Mindspark in November 2017 and endline testing in February 2019, students’ performance in both Math and Hindi were improved across multiple grade levels. Mindspark students scored 0.2 standard deviations more compared to a control group in both the subjects. In other words, learning gains for students that received Mindspark was twice as much as students that did not receive Mindspark. Based on these results, Dr. Shawn Cole (John G McLean Professor of Business Administration at Harvard Business School) also wrote a case study that is now taught to students of Harvard Business School and Harvard Kennedy School of Government.

Student progress and personalised student journey on Mindspark

In 2012, Ei set up Mindspark learning centres for low-income communities in South Delhi, with CSF’s support, which also partly funded the development of Mindspark Math and language products in Hindi. From 2015-2016, a randomised controlled trial was conducted by the researchers of Abdul Latif Jameel Poverty Action Lab (J-PAL), Massachusetts Institute of Technology led by Professor Karthik Muralidharan, (Co-chairperson of Education Research) to test the impact of Mindspark on student test scores in Math and Hindi. The study was conducted among 619 students between grades 6 and from five public middle schools close to the Mindspark centres. The results showed significant and promising improvements of 0.37 standard deviations in Math, improving by over twice as much as students in the comparison group and 0.23 standard deviations in Hindi, improving by 2.4 times as much as students in the comparison group over just a 4.5 month period. The relative impact of the programme was found to be much greater for low-achieving students, who were making no progress in school. The findings stated that these were among the largest gains of any education intervention that were observed in India.

Students using Mindspark in government schools in Almora, Uttarakhand in 2022-23.

In 2023, Ei was selected as a partner on the LiftEd EdTech Accelerator portfolio and continues to work on improving the
efficacy and engagement of Mindspark for early grades.

Strengthening Mindspark’s FLN Content for Early Grades

The LiftEd EdTech Accelerator supports the creation of a robust supply of pedagogically sound and contextually relevant EdTech solutions for the low-income segment in India. As part of the Accelerator portfolio, Ei aims to strengthen FLN content on Mindspark and enhance the adaptive logic used to deliver foundation skills to students.

  • Read information with comprehension to understand scenarios
  • Express meaningful information in oral and written form
  • Understand and present ideas through mathematical symbols and expressions
  • Operate with numbers to reason with numerical scenarios

With these goals in mind, Ei assessed the existing Mindspark FLN content to understand the efficacy of questions, learning modules and the logic of administering the questions to effectively teach concepts. This was done by using student data collected through Mindspark over the last five years to understand the effectiveness of skills covered to improve student learning outcomes and identifying the skills that could be introduced for both literacy and numeracy.

Based on data analysis, new skills were added to the FLN content module which covered essential literacy and numeracy skills in Hindi, employing interactive exercises, visual and auditory stimulus and feedback for students. These were then evaluated through preliminary pilots with 120 students and 11 teachers in two schools in the Bilaspur district of Himachal Pradesh and in the Dungarpur district of Rajasthan, to understand the response of children and teachers to Mindspark content. Since many of these modules existed in the current version of Mindspark, they were improved upon prior to the pilots, and based on the findings from the pilots additional iterations were made to enable a continuous and iterative approach to strengthening FLN modules.

Students and the Mindspark team during preliminary pilots to understand the response of children and teachers to Mindspark content

Findings from the pilots led to product iterations on aspects of interface, content development and pedagogy.

In terms of content development, the feedback and data collected were instrumental in the iterative development of Mindspark’s FLN modules, ensuring they are more closely aligned with the student’s context. Findings from the pilot also highlighted the importance of consistency, interface navigability, UI/UX updates for intuitive response mechanisms for children. Ei also incorporated pedagogical learnings from the pilot which included dialect influences on language learning, sequencing of content, scaffolding of content basis practice requirements, audio instructions, and enhanced personalised delivery of content based on student learning levels.

Iterated questions for Hindi on Mindspark
Iterated questions for Math on Mindspark

After the identified content, interface and pedagogical iterations were made, the modified FLN content developed during the first year of the LiftEd EdTech Accelerator, was deployed to ~10,000 low-income students in Grades 1 to 4 across 13 districts in Bihar, Himachal Pradesh and Rajasthan. The initial response to the launch of updated Mindspark content has been positive, particularly in areas with the lowest learning levels.

Looking Ahead: Ei’s Vision for Year 2 of the LiftEd EdTech Accelerator

In 2024-25, through the LiftEd EdTech Accelerator, Ei is planning to reach 45,000+ low-income learners from ages 4 to 10 through Mindspark. In alignment with the Accelerator’s mission of transforming tech-based learning at home for the low-income segment in India, Ei is cognisant that merely scaling EdTech is not enough —learning is the primary metric for the success of a programme. Ei’s core belief is that usage of Mindspark leads to learning. Hence, it hopes to continue iterating innovative programmes and product features to sustain engagement and usage with an increased user-base for improved student learning.
Over time, Ei aims to build upon the content improvements from the first year of the Accelerator and continuously enhance the adaptive logic of Mindspark. Ei will also enable software-based baseline and endline tests to assess children at their skill level, and measure their learning improvements over the year. Additionally, Ei aims to experiment with the latest advancements in AI to build solutions for FLN in vernacular languages. The progress made in the first year to fortify FLN content has laid a strong foundation for making Mindspark accessible for the last mile of learners.


To learn more about the LiftEd EdTech Accelerator, click here.

[1] ASER 2022 Findings

[2] National Achievement Survey 2021

[3] Winthrop, Rebecca and McGivney, Eileen. “Why wait 100 years? Bridging the gap in global education” Brookings Institution, Available at https://www.brookings.edu/research/why-wait-100-years-bridginfg-the-gap-in-global-education/, Accessed on 19 October 2021.

[4] Misconceptions during the learning process refers to common errors or gaps in knowledge that lead to inaccurate conclusion or logic, such that the student lacks the adequate foundation in that subject.

[5] Teaching at the Right Level (TaRL) is a pedagogical approach for teaching based on the learning levels of a student, learning levels can be understood better through assessment data.