Final answer:
The main challenge in investigating wage discrimination empirically is distinguishing true pre-market characteristics from the effects of discrimination on those variables, leading to potential errors in the analysis of wage differentials.
Step-by-step explanation:
A major challenge that is involved in the empirical investigation of wage discrimination is that some of the control variables are not really pre-market characteristics, as their values might reflect discrimination rather than cause it. This issue makes it difficult to disentangle the effects of discrimination from other variables when analyzing wage differentials. Economists use tools such as multivariate regression to account for observable characteristics like occupation and education when analyzing wage disparities, but unexplained wage differentials can remain, which might be attributed to discrimination. Yet, these residuals are complex as they may also capture effects of discrimination on the variables themselves, such as education, rather than just indicate pure wage discrimination.
Therefore, the challenge with the empirical investigation of wage discrimination lies in the accurate separation of what is truly a pre-market characteristic from the influences of discrimination that have shaped those very characteristics. This means that the Oaxaca decomposition, an economic method used to decompose wage differentials, could be subject to ability bias where it falsely attributes differences to productivity-related factors instead of potential discriminatory practices.