> The idea seems to be that any differences in group outcomes must be eliminated—even if the only way to achieve this goal is to water down the test. Can someb

> The idea seems to be that any differences in group outcomes must be eliminated—even if the only way to achieve this goal is to water down the test. Can somebody explain to me how watering down tests, to lower the competence bar, could possibly address differences in group outcomes? Obviously these different groups don't have different levels or distributions of competence, so how is one meant to relate to the other? > On top of all that, an American Civil Liberties Union representative provided conference attendees with a lecture on criminal-justice reform in which he argued that states should minimize or overlook would-be lawyers’ convictions for various criminal offenses in deciding whether to admit them to the bar. This part I get. Cops are biased and arrest some groups at a higher rate than others. A prior criminal conviction for smoking dope shouldn't have any bearing on the quality of a lawyer, when the competition also smoked dope and simply didn't get caught. This all makes sense. But watering down the tests? What the hell...