Dropout Mitigation: A Logistic Modelling Approach

Working Paper, 2019

This paper seeks to indexing the potential number of dropouts from a particular governmental school in a given year using statistical learning and logistic modelling.

This was an independent research project, carried out when I was a senior in high school.

I worked to execute a complete research project from scratch, and did the following:

  1. Ideate the research, and design what I could study.
  2. Create, conduct, and digitize field surveys (using an iterative feedback mechanism).
  3. Carry out a comprehensive literature review, and incorporate the findings into my research.
  4. Create a logistic model to calculate the propensity of students dropping out from the government school.
  5. Create and distribute free software that helps schools identify ‘at-risk’ students.
  6. Consult with school administrations and parents to mitigate school dropouts.
  7. Document the determinants of dropout risk, and help schools best utilise thier resources to target the improvement of the determinants.

My research aims to perform quantitative analysis on the various factors that influence the number of dropouts from governmental schools across the country to create a ‘dropout index’ which predicts, based on various objective factors, how many students from a government school will drop out in that year.

The paper can be downloaded from this link.