What Drives Success? Determinants of Secondary School Performance in Argentina

In this project, I analyzed data from the Argentine government to explore the factors influencing secondary school students' performance in Argentina. Using the Aprender Test—a national assessment that evaluates student learning and provides insights into their educational conditions. Skills demonstrated: Python, Exploratory Data Analysis (EDA), data collection, data wrangling, data visualization.

11/30/20244 min read

woman wearing blue denim jacket holding book
woman wearing blue denim jacket holding book

Why Educational Analysis Matters in Argentina

1. Ensuring the Right to Quality Education for All

In Argentina, formal education is a fundamental right, yet challenges persist in delivering high-quality education. Recent studies from the Ministerio de Educación de Argentina (2022, Informe de Resultados Aprender 2022) show that average scores among secondary students have declined by over 10% in core subjects like mathematics and reading comprehension.

2.Empowering Youth for a Changing Labor Market

Secondary education is essential not only for students pursuing higher education but also for those preparing to enter the workforce. With youth unemployment rates in Argentina at around 20% (Instituto Nacional de Estadística y Censos, 2023, Informe sobre la Situación Laboral de los Jóvenes), improving educational outcomes equips young people with the skills needed to compete in an evolving job market.

3.Promoting Equity and Social Mobility

Education remains one of the most powerful tools for reducing social inequalities. Studies from Observatorio Argentinos por la Educación (2023, Análisis de la Equidad en la Educación Argentina) indicate that students from lower-income families are disproportionately affected by educational setbacks. By identifying and addressing the factors behind academic outcomes, we can target interventions to help bridge these gaps and promote a more equitable society.

Define and estimate a model that will allow us to compare and explain the performance of students in the last year of secondary school in Argentina in 2022, specifically in the subjects of language and mathematics, considering the heterogeneity of the different regions of the country.

Using statistical analysis, programming, and mixed linear modeling techniques, I uncovered hidden patterns within the data.

These insights allow for the development of algorithms, probability estimations, or statistical models that can support informed decision-making.

The findings highlight key areas where policy and educational interventions could be effectively targeted to enhance academic outcomes for students.

Typically ranges from 3 weeks to 2 months.

The project begins with a thorough assessment of your needs and gathering the relevant data. Once collected, I conduct a detailed analysis and share preliminary insights. Finally, I develop an algorithm or model that you can implement directly to support data-driven decisions and targeted actions.

person writing on brown wooden table near white ceramic mug
person writing on brown wooden table near white ceramic mug

Goal Result Project Duration

Goal

Result

Project Duration

DATA & Variables

The data for this project is sourced from the Aprender 2022 standardized tests for secondary education, provided by the Secretariat of Educational Evaluation within the Ministry of Education, Culture, Science, and Technology of Argentina.

The sample consists of 403,468 observations across 231 variables. To account for the hierarchical structure of the data, students were grouped into five regions of Argentina: Center, NOA (Northwest), NEA (Northeast), Cuyo, and Sur (South).

I chose to employ a mixed linear model, which extends general linear models by incorporating both fixed and random effects. Fixed effects are constant parameters that influence the entire study population, while random effects allow certain parameters to vary across groups or hierarchical levels—in this case, across the diverse regions.

All statistical and econometric analyses were conducted using Python.

Results: Key areas for policy and Educational interventions

The analysis of the Aprender 2022 tests reveals several key areas where educational policies and interventions can be targeted to improve academic outcomes:

Negative Effects on Academic Performance:
Several factors have been identified as negatively impacting student performance, including:


-
Use of Cell Phones with Internet: Excessive use can distract from studies and diminish focus.

-
Paid Work: Balancing work and school responsibilities may limit study time and affect performance.

-
Repetition of Grades: Students who repeat grades often struggle with motivation and academic success.

-
Absenteeism: Frequent absences can disrupt learning and hinder academic progress.

-
Teenage Parenthood: Students who become parents at a young age face additional challenges in completing their education.

-
Living in Rural Areas: Geographic disadvantages can limit access to quality educational resources and support.

-
Attending Public Schools: Students in public schools may face more challenges than those in private institutions.

-
Being a Foreign Student: Immigrant students often encounter barriers related to language and integration.

-
Suffering from Bullying: Bullying can significantly impact a student's mental health and academic engagement.


Factors Associated with Higher Probability of Success:
Conversely, certain student characteristics are linked to a higher probability of success in the Aprender 2022 tests:


-
Higher Socioeconomic Status: Students from families with greater resources tend to perform better academically.

- Parental Education: Students whose parents have completed university studies often benefit from a more supportive learning environment.

-
Urban and Private School Attendance: Students in urban areas and private schools generally have access to more favorable social environments and educational resources.

These findings highlight the importance of targeted interventions aimed at addressing the challenges faced by vulnerable student populations while reinforcing the support systems that contribute to academic success.

Achievements

I am proud to share that I won second place in the poster contest at the LI Argentine Statistics Colloquium, organized by the Argentine Statistical Society. Competing against outstanding posters from across Argentina, I had the honor of representing the Universidad Nacional de Córdoba. This experience not only showcased my research but also allowed me to connect with fellow statisticians and enhance my understanding of the field.

For more details, you can read the article published by my university