Or, at least that is one possible reading of this following example from HESA’s Guidelines for the use of the DLHE Longitudinal Survey Dataset.
To illustrate how this is done:Black, mixed and other ethnic group graduates accounted for 21.9% of the selected Sample A.
From the initial census it is known that these graduates represent just 4.9% of all graduates
To ensure that these graduates feature in the analysis in their correct proportion, the ‘black’, ‘mixed ethnic group’ and ‘other ethnicity’ graduates in the sample would be given a weight of 4.9/21.9.
You mean weight the results by 5/20 because most of the respondents were black! What is the correct proportion? This is a small example of how marginalised groups appear to have to work five times as hard just to be level.