Youth Alcohol Usage and Academic Success Analysis

Dataset Features

• Demographics:
  • school: Student's school (binary: 'GP' - Gabriel Pereira, 'MS' - Mousinho da Silveira).
  • sex: Student's gender ('F' - female, 'M' - male).
  • age: Student's age (15–22 years).
  • address: Home address type ('U' - urban, 'R' - rural).
  • famsize: Family size ('LE3' - 3 or fewer members, 'GT3' - more than 3).
• Family Information:
  • Pstatus: Parents' cohabitation status ('T' - together, 'A' - apart).
  • Medu & Fedu: Education levels of mother and father (0–4, from none to higher education).
  • Mjob & Fjob: Mother's and father's jobs ('teacher', 'health', 'services', 'at_home', or 'other').
  • guardian: Primary guardian ('mother', 'father', 'other').
• Lifestyle and Social Behavior:
  • studytime: Weekly study time (1–4, from <2 to >10 hours).
  • traveltime: Travel time to school (1–4, from <15 min to >1 hour).
  • Dalc & Walc: Alcohol consumption on weekdays and weekends (1–5, very low to very high).
  • goout: Time spent with friends (1–5, very low to very high).
  • romantic: Romantic relationship status (yes/no).
• Academic Performance:
  • G1, G2, G3: Grades for first, second, and final periods (0–20).
• Health and Activities:
  • health: Current health status (1–5, very bad to very good).
  • absences: School absences (0–93).
  • activities, schoolsup, famsup: Participation in extracurricular activities and support programs (yes/no).

Context

Usage:

• Investigate the impact of lifestyle and family background on academic performance.
• Explore how alcohol consumption affects students' grades and behavior.
• Create targeted interventions to reduce alcohol consumption among students.
• Design programs to enhance academic success based on study habits and social factors.
• Examine correlations between alcohol use and health among teenagers.
• Study behavioral trends among urban and rural youth populations.
• Develop machine learning models to predict academic performance based on lifestyle factors.
• Explore data for regression and classification tasks in educational domains.

Who can use it:

• Education Research:
• Policy Development:
• Health Studies:
• Predictive Modeling:

Acknowledgments

• Source: Data provided by P. Cortez and A. Silva, and originally published as part of the FUBUTEC 2008 proceedings.
• Citation: P. Cortez and A. Silva. Using Data Mining to Predict Secondary School Student Performance. FUBUTEC 2008, Porto, Portugal.
• License: Open Access.