Enhancing Operational Efficiency through Data: A Humble Journey with SEF and DataKind Bengaluru

In our previous blog post, we shared the exciting collaboration between DataKind Bengaluru and the Simple Education Foundation (SEF), a project that sought to leverage data-driven insights to identify patterns and trends in educational outcomes, teacher practices, safety levels in classrooms & impact on children’s academic fluency & socio-emotional growth. Today, we present a follow-up post that dives deeper into the expanded tech stack that played a pivotal role in achieving this goal. Additionally, we will highlight the dedicated efforts of our volunteers and their sensitivity towards meeting SEF’s needs by setting up a simple yet impactful solution, while also acknowledging the limitations encountered with certain tech stack elements due to changing needs & processes.

Tech Stack Overview

• Survey Form: Kobo Toolbox — KoBo Toolbox served as the primary data collection tool, enabling the SEF team to gather valuable information directly from the field. With its user-friendly interface and offline capabilities, KoBo Toolbox provided a reliable solution for a diverse range of data collection needs, empowering SEF to make informed decisions.
• Data Preparation (ETL): Python — Python, a versatile programming language, played a crucial role in the Extract, Transform, Load (ETL) processes. It facilitated the seamless extraction of raw data collected through KoBo Toolbox, transforming it into a structured format for further analysis. The volunteers skillfully used Python libraries such as Pandas to cleanse and manipulate the data, ensuring high-quality insights.
• Code Deployment: AWS Lambda — AWS Lambda, a serverless computing service, streamlined the deployment and execution of code. Its cost-efficient and scalable nature enabled real-time data processing, making sure that SEF had access to the latest information at all times.
• Job Scheduler: AWS Cloudwatch Eventbridge — The team relied on AWS Cloudwatch Eventbridge as a dependable job scheduler to automate recurring data processing tasks. This seamless automation reduced manual intervention and ensured a smooth and efficient data pipeline.
• Data Visualization: Looker Studio — To make the collected data more accessible and actionable, the team utilized Looker Studio for visualization purposes. These platforms allowed the creation of intuitive and interactive dashboards, helping SEF stakeholders gain valuable insights to drive their initiatives.

Shortcomings of Using KoBo Toolbox as a Data Collection Tool

While KoBo Toolbox offered valuable features, it did present some challenges during the data collection phase:

• Limited Customization: KoBo Toolbox offers predefined question types and layout options, limiting the scope of customization for survey forms. In cases where specific data points needed to be captured, the tool’s constraints posed challenges, requiring workarounds or compromises.
• Learning Curve: While KoBo Toolbox is relatively easy to use, there was still a learning curve for some team members unfamiliar with the platform. Training and familiarization sessions were necessary to ensure the accurate deployment of the survey forms and to address any potential errors.

Challenges in Changing Data Structures from Google Forms to KoBo Collect

During the migration of data collection from Google Forms data stored on Google Sheets to KoBo Collect, the project team encountered several difficulties:

• Data Mapping: As the data structures in Google Sheets and KoBo Collect were different, mapping the existing data fields to their corresponding ones in KoBo Collect posed a significant challenge. This process required careful attention to detail to avoid data misalignment and inconsistencies.
• Data Validation: Validation rules and data constraints differed between Google Sheets and KoBo Collect. Ensuring that data captured through KoBo Collect adhered to the same standards as the previous system demanded thorough testing and validation.
• Error in Looker Studio dashboards: We found that due to changes in the final processed data, the graphs developed early on based on the Google forms were breaking without helpful error messages. With experience, we came to learn the different settings to check when changing a data source.
• Stakeholder Communication: Managing expectations and communicating changes to all stakeholders was crucial to avoid misunderstandings and confusion. Transparent communication was essential to gain buy-in and support for the migration process.

Embracing Humble Progress towards evidence backed work inside government classrooms

Rather than describing the impact as “tremendous,” we humbly acknowledge the steps taken towards enhancing operational efficiency at SEF. By harnessing the power of data and technology, the collaboration between DataKind Bengaluru and SEF has laid the foundation for more informed decision-making and improved processes.

The Role of Dedicated Volunteers

Central to the success of this initiative are the dedicated volunteers Amrita & Abhiram from DataKind Bengaluru. Their unwavering commitment and passion for social impact have been instrumental in driving the project forward. With the Learning & Insights Team at SEF (Shadab, Srinidhi, Kabir) support in helping us understand SEF’s mission, our volunteers worked tirelessly to accommodate the organization’s needs and challenges.

Sensitivity in Solution Design

Our volunteers carefully tailored the solutions to cater to SEF’s specific requirements. By listening to SEF’s stakeholders and incorporating their feedback, the team created user-friendly dashboards and reports that were easy to interpret and act upon.

Conclusion

The collaboration between DataKind Bengaluru and the Simple Education Foundation exemplifies the positive outcomes that can be achieved through data-driven initiatives. By adopting a humble approach to progress, we have taken significant strides in increasing SEF’s operational efficiency.

We extend our heartfelt gratitude to Amrita and Abhiram whose dedication and sensitivity to SEF’s needs have been invaluable in this journey. Together, we continue to explore the potential of data and technology in addressing social impact challenges. As we move forward, we remain committed to driving meaningful change through data, one step at a time.