3 ms·
Direct source: https://cued.uic.edu/wp-content/uploads/sites/219/2024/05/Handling-Hardship-Final_May2024.pdf https://cued.uic.edu/wp-content/uploads/sites/219/2
by ars 2y ago
Direct source: https://cued.uic.edu/wp-content/uploads/sites/219/2024/05/Handling-Hardship-Final_May2024.pdf https://cued.uic.edu/wp-content/uploads/sites/219/2024/05/Ha...
"They were recruited into the survey using the Meta/Facebook" that's a red flag on the quality of the study.
"To provide a more demographically representative picture, we reweighted survey responses using data on race and gender composition that Amazon reported to the EEOC in 2021.2" also a red flag because they already don't necessarily have a representative sample, and then are reweighting using data from a different selection criteria.
The study said that 13% of workers moved into higher paying jobs within Amazon, and their numbers were removed from the study.
Yah, I think I'm going to stop reading this study. Very low quality work.
- taeric 2y agoRemoving those that progressed out of the pay scale feels especially odd? That should be seen as a success of employment for this, no? I'm also not a fan of focusing the headline or finding on a single company without a cohort to compare against. If anything, this would punish individual companies for hiring low income people at all. Easy to have 0% here if you just refuse to hire anyone that qualifies, after all. And that is a terrible goal. If I'm off on any of that, happy to be corrected.