Background

Would you like to use your skills in statistics and data analysis to contribute to a better understanding of real-world traffic crashes and injury prevention?

In-depth crash databases provide valuable information about how and why crashes occur and how injuries can be prevented. However, these databases may contain sampling biases, which need to be accounted for to ensure that results accurately represent the relevant country or region.

The Road Accident Sampling System India (RASSI) is an in-depth crash database containing data from more than 8,000 crashes collected over a 12-year period. RASSI currently provides weighting factors to adjust for sampling bias at crash level, but there is a need to extend this methodology to person and injury levels.

Another international database, the Initiative for the Global Harmonisation of Accident Data (IGLAD), currently lacks weighting factors, which limits how the data can be used in representative analyses.

About the thesis

In this Master Thesis project, you will explore and compare statistical methods for correcting sampling bias in in-depth crash databases.

The main focus will be to develop multilevel weighting factors that can be applied from crash level down to person and injury level. You will evaluate different weighting approaches and identify a suitable method for the application.

The selected method will then be applied to IGLAD data to explore whether the methodology can be transferred to another international crash database and to assess its strengths and limitations.

Your thesis will include

• Evaluating and comparing different weighting and bias-adjustment methods
• Developing multilevel weighting factors for RASSI, from crash to injury level
• Comparing the developed weighting factors with existing RASSI weighting factors
• Applying the selected method to IGLAD data
• Evaluating and highlighting the strengths and limitations of the selected approach

Who are we looking for?

We are looking for 1–2 Master’s students with an interest in statistics, data analysis and real-world applications.

You should have:

• Knowledge of statistics and data analysis
• Experience with R programming (preferred) or Python
• Knowledge of traffic safety epidemiology and sample design is beneficial but not required.

Interested?

For more information about the thesis, please contact:

• Hanna Jeppsson hanna.jeppsson@autoliv.com
• Junaid Shaikh junaid.shaikh@autoliv.com
• Jordanka Kovaceva, Chalmers University of Technology jordanka.kovaceva@chalmers.se

More lives saved – more life lived!