Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

How can we further remove biased from this study?

I agree this looks impressive i the level of research and detail, but I strongly caution the community not to jump to conclusions.

Be mindful of where the data comes from. Police and prosecutors can and do make mistakes, inadvertently or sometimes prejudiciously.

I've heard this statistic most of my life that Black people commit a significant amount of crime, yet being such a small percentage of the population. I just think it grossly over simplifies everything and its unfair.

I'm not sure how to account for that bias, I don't think we can draw conclusions until we can rectify some of that



> Police and prosecutors can and do make mistakes, inadvertently or sometimes prejudiciously.

That's one of the reasons murder is used. Hard to fudge the statistics on it, immune to overpolicing, most of it is intra-racial, and it's hard to misidentify a perpetrator so bad that they go from white to black.

But for non-murder, there is the National Crime Victimization Survey, that does not involve the police, and acts as a check on their data: https://en.wikipedia.org/wiki/National_Crime_Victimization_S...




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: