Introduction
In 2021, India's National Achievement Survey (NAS) assessed nearly 3.4 million students to understand the nation's state of educational achievement. Before that, in 2017, 2 million students were assessed under the same program. Govt. of India has released data from 2017 and 2021 assessments for the public (see public report cards). The government-provided report cards do not show how things changed from 2017 to 2021. They also lack the actionable information that many stakeholders want. We decided to turn things around and created actionable data dashboards with longitudinal data. In this post, we will walk through a new National Achievement Survey Data Dashboard that shows the critical information sought by stakeholders. The dashboard has been developed by data scientists at Playpower Labs.
Government-provided report cards available at https://nas.gov.in/report-card
Purpose
Education is the cornerstone of a nation's growth and progress. The government of India foresees education as a cornerstone for the economic growth of the country. The National Assessment Survey (NAS) refers to a standardized assessment test that is conducted by the government or a governing body of a country to evaluate the education system's effectiveness and identify areas that require improvement.
The purpose of the NAS is to provide policymakers, educators, and other stakeholders with valuable insights into the strengths and weaknesses of the education system, including curriculum, teaching methodologies, and learning outcomes.
This information can be used to inform policy decisions, allocate resources, and develop interventions to improve the education system's overall performance.
Motivation
Although extensive surveys were conducted in 2017 and 2021, with the resulting data being in the public domain, we have decided to create a dashboard that compares the data from both years. Public dashboards and datasets provided by the government are not longitudinal and do not aid in making better policy decisions by providing feedback to stakeholders on the effects of policy changes.
Methodology
We used the Python Requests and Beautiful Soup library to perform data scraping. The dataset for 2017 was readily available in the form of PDFs, which we downloaded for specific districts and extracted the data using scraping techniques. For the 2021 dataset, the information was in a JSON format on the NAS website, and we used Python Requests and the JSON library to extract the information. Both datasets were converted into CSV format, and we used Shiny Dashboard to create the dashboard.
Analytics
To access the new NAS data dashboard developed by us, click on the following link:
The dashboard was created with the practitioner's perspective in mind. It includes the following dashboards:
1. Score Improvements (link)
This dashboard provides an overview of the improvement in the overall rankings based on all the subjects. Users can select various differentiators, such as males, females, rural, and urban.
2. Rank Improvements (link)
This dashboard is similar to the improvements dashboard, but in this dashboard, the user can select various subjects as well.
3. Scores Analysis, Rural versus Urban (link)
This dashboard focuses on state-wise scores based on the various subjects. A specialized feature for rural dominance is also provided so that the performance of the rural population can be seen alone and not get masked due to the high scores of the urban population. The proficiency level of the students could be looked at using this dashboard.
4. State Drill Down (link)
This dashboard gives both an overall view and a district-wise ranking of the state districts. Male, female, urban, and rural populations can also granularize the data.
5. District Comparison (link)
This dashboard compares two districts across states or within the same state. The comparison also shows both the data for 2017 and 2021.
6. Grade Level Comparison (link)
In this dashboard, the user can select a district and compare districts across different grades. Additionally, this dashboard also gives the user the ability to check correlations across males and females or urban versus rural as well.
7. Feedback Dashboard (link)
This dashboard presents the qualitative feedback taken by the surveyors as a percentage of the total.
8. LOs Dashboards (link)
This dashboard allows the user to map the learning outcomes with the percentage of scores. This could be done for a specific subject or overall subjects.
9. Correlations (link)
This dashboard allows the user to check correlations between feedback and the score for a particular subject or for all the subjects in a particular grade.
Live Dashboard (Preview)
Below you can access the live dashboard. It is recommended to open the dashboard in a new window using the direct link.
Suggestions
We welcome any comments or feedback on the dashboard. If you feel that any perspective is missing, please feel free to share your thoughts with us. We are always open to incorporating changes to improve the usability of the dashboard.
If you have any technical questions regarding the data scraping process or the combined data, you can reach out to us at the email address provided gyanesh.jain@playpower.com. Alternatively, you can wait for our next blog post where we will be sharing more technical details about the dashboard.
Thank you for your interest in our dashboard, and we look forward to hearing from you.
About the Author
Gyanesh Jain is a senior data scientist at Playpower Labs. I have done my Ph.D. in predictive Analytics from the University of Lucknow. I have done my Post Graduate Diploma in Business Analytics and Intelligence. I am interested in creating and utilizing technology to ease education to the lowest strata of society. My other interests include econometrics, time series forecasting, and reinforcement learning. I believe in constant learning. Love to hear from you in the comments. Please feel free to connect via LinkedIn.
Comments