So larger populations and smaller number of attributes would probably make it better, but not foolproof.
Seems pretty useless without that. Those are huge factors.
"The question is mainly whether you want to be ignorant, or not."
That's not the only question. The other question mostly being discussed here is if you want anyone to be able to know your salary by working backwards from the published list, specifically in small companies.
The point is to allow for more transparency, not perfect transparency.
> That's not the only question. The other question mostly being discussed here is if you want anyone to be able to know your salary by working backwards from the published list, specifically in small companies.
In practice this doesn't matter, but even if this were an issue you could exempt employers with fewer than 100 employees.
How so? Even at 100 employees you might only have 5 tech leads. So the population of the subgroups matter.
"The point is to allow for more transparency, not perfect transparency."
What can we actually do with that transparency if it doesn't have the attributes necessary for meaningful comparison?
Are you arguing knowing nothing is better than knowing something?
No one is saying that. That is a gross misinterpretation of my statements.
"The point is to create a range in which both current and prospective employees can refer to."
A range based on what? The point being discussed here is equal pay. Ranges dont help if you have all the women at the low end and the men at the high end, hypothetically. If you just want to post salary ranges by position you can do that without publishing individual salaries.
Even with that said, I work at a small enough company where anonymous salaries could still easily be associated with an individual, and I would still not like that, for privacy reasons.