Assessment of non-linear dimension reduction methods for calculating a SEIFA-like index

Overview

This project would involve applying one or more non-linear dimension reduction methods to Census 2021 data, to calculate an index similar to the Index of Relative Socio-economic Advantage and Disadvantage (IRSAD) from the Socio-Economic Indexes For Areas (SEIFA). Some information and recommendations for the specific non-linear dimension reduction methods will be supplied as the output from an earlier project.

The methods will aim to measure the same underlying concept as IRSAD and will probably use the same starting set of candidate variables.  The data that was used to create the 2021 IRSAD can be supplied for this research, along with information about the variables.

The results provided by the non-linear dimension reduction methods will be compared against the IRSAD scores.  An assessment of any key differences between the results will be of particular interest – this could include a comparison with real-life data sources and an analysis of the methodological causes of the differences.  The report should also include commentary on the complexity of applying the alternative methods.

Research engagement

Critical Literature Review, Computational Implementation, Quantitative Assessment

Research activities

Comparison of existing methods, literature review, computational analysis

Research skills

Computational Capabilities, Critical Thinking, Written Communication

Outcomes

Written report and recommendations to ABS, possible presentation to ABS

Skills and experience

Statistical and Data Science Literacy and skills preferred

Start date

2 November, 2026

End date

19 February, 2027

Location

GP Y Block Level 8

Keywords

• Non-linear Features
• Data Analysis

Contact

Gentry White

3138 1658

whiteg5@qut.edu.au