MST2074 R-Based Gender Income Analysis Assignment: Investigating Inequality Across Gender and Indigenous Status in Australia

This assignment is an individual work that requires empirical quantitative analysis using the R program. The task will train you to develop various models and explore which model would be the most robust and usable to solve the problem.This assignment is associated with SDG 5 (Gender equality), including other societal discrimination on gender, region, and indigenous status. Note that there are no fixed answers, and reasonable marks will be rewarded based on the depth of the model development. The model development process contains all the possible techniques that you have learned in this unit, such as variable transformation, interaction terms, functional forms, and handling HSK.

The data seems rather simple, but you can come up with many variations of regression models. Note that the report of your results must be no more than 7 pages (this excludes the reference section, but the page limit includes ample space for tables and figures). Marking will focus on the clarity and conciseness of your presentation; overly wordy or detailed submissions will be penalised.

The issue of gender discrimination in the labour market leads to passionate debate. You are asked to investigate this issue in Australia and write a short report detailing your findings.

You have some data for Australia. The data records incomes for different segments of the population. The population is segmented by several variables (region, Indigenous status, employment status, English language proficiency, and sex). For each of these segments, we have the average age, income, and schooling. You are to use this data to undertake an investigation of the extent to which there are differences in incomes across genders and across indigenous/non-indigenous people in Australia. In particular, consider the following quotes from Beyoncé and Serena Williams, respectively, regarding the US:

Compare your findings with those in the US, according to Beyoncé and Serena Williams. An important feature of the data is that observations represent average values across individuals and not observations on single individuals. This has implications for estimation.

Coding: All the working and output must be stored and organised (i.e., use # to label what your codes attempt). The codes should not be included in the report. All the results and relevant output should be included in your report. (5 marks)

Introduction: Your introduction should briefly outline the problem and discuss the nature and some of the relevant features of your data for your investigation (summary statistics and visualisations). Here, you may particularly want to compare male vs female incomes and indigenous vs non-indigenous incomes. (10 marks)

Modelling incomes in Australia: Construct a regression model of incomes in Australia and discuss the process you used in model development (e.g. choice of variables, significance tests of variables, functional form, weighting). Do not forget to perform any relevant statistical tests to confirm your findings. Then, interpret the coefficients and discuss whether there is evidence of gender differences in incomes and differences across Indigenous/non-indigenous people. (25 marks)

Conclusion: Briefly summarise what you have found about the drivers of income in Australia. Outline some of the major limitations of your analysis. What aspects of the data and/or variables and/or modelling may limit the prescriptive/causal interpretation of your results? (10 marks)

References: Please include any references you used for the report.