4 ms·
This will end up biased by the largest employers. Kind of like crime maps--they make it seems like there's an extraordinary amount of crime concentrated in urba
by jimmar 2y ago
This will end up biased by the largest employers. Kind of like crime maps--they make it seems like there's an extraordinary amount of crime concentrated in urban areas, but that is due (in part) because that's where a lot of people live. A small company might ghost 90% of applicants and have a low count of ghostings, whereas a large employer might ghost 2% of applicants and have a very large count of ghostings.
This website could be improved by recoding employers who did not ghost applicants.
- zooq_ai 2y agoHa Ha, all maps are population maps
- drewcoo 2y agohttps://repository.hou.usra.edu/handle/20.500.11753/1254 https://repository.hou.usra.edu/handle/20.500.11753/1254
- cxr 2y agoSo true. Another example: I would say I murder less than 1% of the people I meet, but do I meet a lot of people compared to the average person? Let's say we tallied up the total number of people I murdered this year and it came out to, I dunno, something in the neighborhood of 4. We can say it's no more than 4 just for argument's sake. That's 4 murders out of hundreds. Is that supposed to be bad?
- figassis 2y agoRank by ghost/headcount
- 0xfedcafe 2y agoWhat are the issues with using a field that has a "specific" normalized ghost rate with ghosting by 1,000 hires?
- Redoubts 2y agoThis is a really funny analogy, because crime maps within a city are very valuable. https://x.com/gothamist/status/1816609920202019080 https://x.com/gothamist/status/1816609920202019080 > Just 4% of New York City’s 120,000 blocks account for nearly all of the city’s shootings, according to the data.
- fsckboy 2y agojust to add on, it is routine to hear crime statistics reported as "crime rates" it's still true that crime rates might be higher when people are close to each other because "rural serial killers need comparatively a lot more travel time," but still, the idea of weighing crime statistics to make statistical sense of them is pretty well established.