Nov 13, 2021 01:04 PM
(This post was last modified: Nov 13, 2021 06:49 PM by confused2.
Edit Reason: Maths fail
)
I get the impression that the root of 'this' is a group with access to the health and employment records of the entire population of Denmark - about 5.8 million people. Data mining is turning up some interesting and some possibly spurious connections. One might expect about half the population of Denmark to be working so a dataset of 250,000 is about 5% [edit 10%] of the working population. Why select only 5% [edit 10%]- the companies with a high incidence of reported mental health problems? Within the dataset there will be companies where the incidence of mental health problems (statistically) rises and falls. Within that subset there will be companies where the incidence of mental health problems rises with the transfer of an employee from another company with mental health problems and you have a clickbait result.
