EDPS 36100 Lecture Notes - Lecture 3: Base Rate
Document Summary
While scholars often disagree about bias in testing- this testing is often related to lack of clarity between terms and usage. If i lower required scores (or raise my selection rate) Success ratio: proportion of selected that succeed. Anytime you alter the selection criterion for any reason, you are going to have some effect on the success ratios. Different impacts of selection vs non selection will be involved in the process. A test is biased for members of a subgroup of the population if, in the prediction of a criterion for which the test is designed, consistent non-zero errors of prediction are made for members of the subgroup. Test bias refers to over-prediction or under-prediction of criterion measures. Darlington argued that fairness can only be achieved by: And allowed for conflation between test bias, test fairness, and fairness in the use of tests/selection processes.