Remix.run Logo
Remix clone Hacker News
new | show | ask | jobsGithub
▲ wavemode 3 days ago

> So a Gaussian distribution is a far more reasonable assumption than a slice of the Pareto distribution

It's not an assumption. See the evidence referenced in the footnotes.

▲crazygringo 3 days ago | parent [-]

Incorrect.

It is absolutely an assumption. The "evidence" in the footnotes is about national salary data. Not the distribution for any individual position at a company.

And it is entirely possible (and probable) that performance at each position is distributed as a Gaussian, and all those Gaussians add up to a Pareto at a population level.

But you simply cannot take national-level data and assume it applies at the micro level. That's not how statistics works.

▲wavemode 3 days ago | parent [-]

> And it is entirely possible (and probable) that performance at each position is distributed as a Gaussian

Find me any research agreeing with this statement

(spoiler - you won't, because researchers have broadly reached consensus to the opposite)

▲crazygringo 2 days ago | parent [-]

OK, can you point us to that consensus please? Anything published?

I would have thought that if there were literature on that consensus, the author would have cited it. But they didn't. And the entire tone of their article is speculative.

Gaussians are generally considered the default assumption for a process with an error term with unrestricted movement in both directions, until shown otherwise.

▲wavemode 2 days ago | parent [-]

> OK, can you point us to that consensus please? Anything published?

Here's just one article - it references numerous studies and meta-studies, broadly making the case that Pareto distributions tend to fit real-world data better than Gaussian (including employee performance, if you observe the section titled "Managing People"): https://hbr.org/2022/01/we-need-to-let-go-of-the-bell-curve