- Team Members: Agarwal Shruti Hemant, Esha Sharma, Sharma Radhika Shivkumar, Sandhu PrableenKaur CharanjitSingh, Patel Vibhuti Kishorbhai, Sagar Khushi Rajeshkumar
- Contact Email: [email protected]
Education is universally recognized as a fundamental human right and a potent catalyst for social mobility and the overall progress of a country. Education is key to creating a level playing field. The dropout rate is a metric that quantifies the percentage of students who discontinue their education before completing a particular level or grade. It serves as a crucial gauge of the education system's effectiveness.
Various social and economic factors lead to students dropping out of school. There has been a significant decrease in the dropout rate in the primary level but a lot more work needs to be done at the secondary and senior secondary levels. Identification of major challenges faced by students can help in the formulation of targeted interventions that can reduce the dropout rate.
Analysis of student dropout rate at the school level based on the following categories:
- Gender
- Area
- Age/standard
- Social Category Understanding overall trends of dropout rate at various levels. Identification of factors that lead to students dropping out of school. The aim is to help in developing data-driven policies for the reduction of student dropout rates.
Drop rate is an important indicator of educational success and can vary significantly by region, country, and socioeconomic factors. It indicates that a significant proportion of students are not completing their education, which can have long-term consequences on their future opportunities and economic prospects. Students who drop out struggle to find better employment. They face a higher risk of child marriage and getting into illegal activities. Thus, it is very important to ensure that more and more students complete their education. According to National Education Policy 2020, the government aims to decrease the dropout rate to zero and to ensure that the gross enrollment ratio in higher secondary schools is 100%.
- Programming Language: Python
- Libraries: NumPy, Pandas, Sci-kit learn, Matplotlib, Seaborn, Plotly
- Tools: Tableau
https://docs.google.com/document/d/1Tg3QbN-kw9wml6TRrSiJ7FsfCiY_P46TpYLMpbESO14/edit
https://public.tableau.com/app/profile/shruti6130/viz/Voyagers-Dashboard/Home#1
- Data has been collected from official Indian government websites and databases.
- data.gov: https://data.gov.in/
- UDISEplus: https://www.udiseplus.gov.in/#/home
- Ministry of Statistics and Programme Implementation, Key Indicators of Household Social Consumption on Education in India (2017-18): https://mospi.gov.in/sites/default/files/publication_reports//KI_Education_75th_Final.pdf
Created a diverse range of visualizations including bar charts, pie charts, stacked bar charts, grouped bar charts, line graphs to understand dropout rates. We focus on national-level, and state-level disparities, age/standard groups, social categories, and gender differences. By creating visualizations like choropleth maps and bar charts, we identify variations among states and regions, revealing critical insights into dropout trends. This multifaceted approach allows us to discern patterns and disparities, enabling more targeted interventions to address dropout challenges effectively.
Utilizing diverse visualizations, our analysis revealed disparities in dropout rates. Recommendations include tailored interventions at the national level, state level, and community engagement efforts.These suggestions come from what the data tells us, to better reduce dropouts and make education fair for everyone.
Applied linear regression model using scikit- learn library to predict the dropout rates for the primary, upper primary, and secondary education levels over the next 10 years with the purpose of using it for comparison with actual trends.
i. Challenges Faced:
- Difficulty in sourcing relevant data.
- Addressing missing values in datasets.
- Identifying key factors for trend analysis.
- Extensive research for effective dropout rate reduction strategies.
- Insufficient data for developing early warning ML models.
ii. Data Search:
- Invested significant time in searching government websites and reports for data.
iii. Visualization Tools:
- Employed Tableau and Python code for diverse data visualizations.
iv. Data-Driven Recommendations:
- Recommendations for reducing dropout rates were substantiated by data analysis.
Overall trends for dropout rate indicate that while there is a significant decrease in students dropping out at the primary level a lot more needs to be done at secondary level education. One positive trend is that the female dropout rate has been lesser than males. The most affected group includes social categories like ST, SC, OBC, and children of migrant workers. The top cause reported by students dropping out of school is that they are involved in domestic work ( for females)/ economic work (for males) followed by financial constraints and lack of interest. In rural areas, one of the primary factors leading to dropout among females is the considerable distance to school. In contrast, males and females in both rural and urban areas tend to leave school due to academic failure. On top of these challenges, migrant workers' kids also face problems due to constantly changing schools.
Top causes stated by students for not attending school, according to the Ministry of Statistics and Programme Implementation (2018) :
- Girls - Domestic work, financial constraints, not interested, marriage
- Boys - Economic work, financial constraints, not interested
For more details refer to the report.
For the reduction of the dropout rate,
- Improvement in the school infrastructure, quality of faculty,
- Financial support for students with financial constraints
- Relevant curriculum and vocational training
- Open Source community for content translation in Regional Language
- Mobile first approach for digital learning platforms
- Student Help group for struggling students
- Early warning system
- National Student ID for keeping track of education progress, especially for migrant worker's kids
- Early warning system for dropouts consisting of reporting by other students and ML techniques.
- Developing prototype for National Studnet ID
| Team Member | GitHub | ||
|---|---|---|---|
| Shruti Agarwal | https://www.linkedin.com/in/shruti-agarwal-2a0019186/ | [email protected] | https://github.com/shrutihagarwal |
| Esha Mishra | https://www.linkedin.com/in/eshamishra28/ | [email protected] | https://github.com/emish8/ |
| Radhika Sharma | https://www.linkedin.com/in/radhika-sharma-476aa41b0/ | [email protected] | https://github.com/Radhika2408 |
| PrableenKaur Sandhu | https://www.linkedin.com/in/prableenkaursandhu | [email protected] | https://github.com/PrableenKaurSandhu |
| Vibhuti Patel | https://linkedin.com/in/vibhuti-patel-5357b920a | [email protected] | https://github.com/Patelvibhuti |
| Sagar Khushi Rajeshkumar | https://www.linkedin.com/in/sagar-khushi-673956232/ | [email protected] | https://github.com/KHUSHIS12 |