Master Thesis – Crash Data Analysis & Weighting Methods

Detailed traffic crash databases support research into crash causes and injury prevention, but sampling bias can limit how well they represent a country or region. RASSI contains records from over 8,000 crashes collected across 12 years and currently offers crash-level weights, while the international IGLAD database has no weighting factors. This thesis at Autoliv Sweden aims to extend bias correction to people and injuries and investigate whether the approach can be transferred between databases.

The assignment

• Review and compare statistical weighting techniques and methods for adjusting sampling bias.
• Develop multilevel weights for RASSI covering crashes, people and injuries.
• Compare the new weights against the crash-level weights already available in RASSI.
• Test the chosen method on IGLAD to assess its transferability.
• Analyse the approach’s strengths and limitations.

Who we are looking for

• One or two master’s students interested in statistics, data analysis and practical applications.
• Knowledge of statistical methods and data analysis.
• Programming experience in R, preferably, or Python.
• Familiarity with traffic safety epidemiology and sampling design is an advantage, but not essential.

Practical details

Location: Vårgårda. The project is intended for 1–2 master’s students. Start date, duration and application deadline are not specified. For further details, contact Hanna Jeppsson (hanna.jeppsson@autoliv.com), Junaid Shaikh (junaid.shaikh@autoliv.com) or Jordanka Kovaceva at Chalmers University of Technology (jordanka.kovaceva@chalmers.se).