Disparity and Discrimination

One of the main problems we often see with discussions of ‘discrimination’ is that it is assumed to be the cause of disparity (differences in outcomes between groups). For example, we see many claims of unequal outcomes (men having higher salaries than women, more men than women in STEM subjects, better outcomes for white people etc.) in society, and these are often attributed to ‘discrimination’ against women, blacks etc.

However as any scientist knows (and as any psychology graduate should know – if they’ve got their BPS accredited degree), we cannot equate correlation (the fact that there may be a relationship between discrimination and a certain outcome) with causation (the idea that discrimination caused a specific outcome). Indeed, it is wrong of us to even imply that a given outcome could be solely due to a given cause when there are any number of alternative explanations. We need to get rid of these alternative explanations before we plump for any single one as an explanation for an outcome.

However, this is what seems to happen with remarkable regularity in the social sciences with psychologists trying to attribute any number of outcomes to discrimination. To illustrate, we regularly see the claim that there is a ‘gender wage gap’ with women being paid less than men, and it is also often attributed to some form of ‘discrimination’ or negativity towards women (or preference towards men). However, when factors such as hours worked, different roles, and different qualifications are taken into account, the gap narrows to cents/pence. While arguably the difference, though minor, is still wrong, there are other potential explanations for it – such as those mentioned in this article discussing a woman’s failed attempt to sue her institution for gender discrimination.

This is of course not to say that differences between groups should not be investigated, but they should be investigated scientifically based on facts. They should not be addressed by people who should know better than to attempt to attribute a false ideological explanation to what is often a misrepresented phenomenon to begin with (and most certainly not one that can be explained by a number of legitimate causes).

For more discussions of the problem of equating disparity and discrimination, see:

• How we think about disparity and what we get wrong

• Commission on Race and Ethnic Disparities report

• Discriminations and Disparities with Thomas Sowell

• Statistical disparities among groups are not proof of discrimination

• Jonathan Haidt briefly mentions the correlation/causation problem but it’s worth looking at its role in the context of critical social justice ideology

• The DDU response to the Runnymede Trust report which attempts to claim that Britain is ‘systemically racist’

• DDU’s response to the Universities UK report

• DDU’s response to the Committee on Racial and Ethnic Disparities’ Report

• The problem with reporting ‘ethnic pay gaps’