This project aims to investigate the proficiency of propensity score matching and its uses in evaluating the gender pay gap. Secondary data is analysed RStudio and propensity score matching is used to produce a balanced data set of 135 males and 135 females across 3 covariates. These covariates included race description, department, and performance score. A t-test is then performed to investigate the null hypothesis: that the population mean salary of males is equal to the population mean salary of females. The t-test showed that there was enough evidence to reject the null hypothesis at a 5% significance, and it was concluded that the population mean salary for males is not equal to the population mean salary for females. This conclusion supports the claim that the gender pay gap is still problematic in modern America.
To make a conclusion on the proficiency of propensity score matching, further tests on larger data sets would be essential, as from this investigation it is unclear whether the matching process increased the reliability of the results, in comparison to a test where propensity score matching is not utilised.