Hacker News Reader

Comment by wdewind | Hacker News Reader

Self reported salaries are just never going to be a good data source. I was hoping since this was LinkedIn and they have tons of resources they'd be doing something different, but they don't seem to have made any progress on the many attempts people make each year to do salary analysis in this way. Anyone have any good ideas for actually gathering accurate salary data?

View on HN

The Federal Statistical Office of Switzerland sends compulsory salary surveys to the enterprises itself. The collected data is then published in the Salarium [1]. It is based on more than 750'000 salaries from the private sector and runs since 1994. You can read more about the methodology in the PDF [2]

I worked as a Software Developer for a company that used the data as base for their salary negotiations and it worked quite well.

[1] https://www.bfs.admin.ch/bfs/en/home/statistics/work-income/...

[2] https://www.bfs.admin.ch/bfsstatic/dam/assets/6937/master

Interesting video about somewhat public tax returns, and hence salary, in Norway and other Nordic countries.

https://www.youtube.com/watch?v=1bO8zEaSuWg

There was no "hence salary" part. The salary back-estimation was very rough, e.g. for my income the estimation by a popular calculator was off by 25%. Using past tense here because anonymous access to tax data is discontinued.

You could tell if someone was poor or rich based on that, but for comparing incomes in your cohort it was not very useful.

IT Jobs Watch surveys recruiters for permanent roles data and contractors quotes for rates.

http://www.itjobswatch.co.uk/

I have used this site. It seems fairly accurate - does everyone else find this too?

For tech work, you can use H1B salary data (from somewhere like http://h1bdata.info/) as a minimum level for a position. Scale up by some percentage (10-20% in my experience) to get the median for what US citizens are likely making in the same position.

The problem is that this only includes base salary.

Yes, you're right. But base salary is the only reliable number to use for comparisons. If we get into things like startup stock options, value is effectively zero on a probability scale. Monetary bonuses are paid out based on all kinds of factors that may or may not come to pass, and have their own taxes associated with them. The only reliable extra comp I can think of would be RSUs and similar at large tech companies, but those are already fairly well known.

In my industry bonuses are as reliable as public company RSUs. To my knowledge, there is less public information on RSU grants and bonuses so I am puzzled by your last sentence.

Where have you looked? For example, a quick search on Quora will tell you the range of RSUs handed out to employees at Google at various levels.

What incentive do people have to lie about their salary (assuming it's anonymous)?

Well in this specific case LinkedIn makes it so you have to report your salary to see other salary data, so that right there is one place where people have an incentive to add junk just to get the data.

But the more important problem is the selection bias.

There are massive incentives to lie about how much you make. It's not hard to imagine this tool will be used by HR compensation experts to determine wage parity. With that in mind, the higher the average, the better positioned you will be to negotiate a higher salary, therefore everyone who reports how much they make is ultimately incentivized to state more than they make in hopes of driving up the wage index for their field.

And who is to stop a software company from doing the opposite? If they want lower wages, they could spam a region with fake workers with lower wages.

Thus to the original point: self reported data is questionable.

It would be neat if the IRS / CRA / national tax body would publish summaries based on people's actual tax returns. Although you don't specify things like industry, and other career specific quirks on your tax returns. (i.e. Ruby on Rails developer vs Dynamics CRM developer)

one of the engineers from Salary team. we talk about privacy and accuracy at length in this blog post https://engineering.linkedin.com/blog/2016/10/bringing-salar.... TL;DR; we do outlier detection.

What's an outlier? I bet a lot of people feel they're shortchanged by about $10k, and will report that much higher to set a baseline. Outlier sounds more like "AngularJS developer making $250,000 in Cleveland"

it means more accurate baseline calculation as well as median salary which will rule out things like "AngularJS developer making $250,000 in Cleveland"

This is the first salary listing I've seen that actually looks somewhat accurate (at least for some titles) I am confused why you guys aren't using the title standardization. You're surfacing and getting different results for "Senior Software Development Engineer" and "Senior Software Engineer" (whatever those mean). You should talk to someone in SNA about this. I think Qi is the guy now.

Also you're having sampling bias that's effecting levels. There's no way the median salary for a staff engineer is higher than a senior staff's. Again talk to Qi.

Stay cool.

Why didn't you buy industry standard employer wage reported surveys like any employer of size does? They are readily available and the cost would be a drop in a bucket for LNKD.

Because they'd rather make and sell their own, obviously

Well, if everybody else thinks that programmers make $200K/year, there's a chance that somebody might consider paying $200K/year. That seems like an incentive to me.

Maybe, but you would need massive collusion with millions of developers that you have never met in order to pull this stunt.

I don't think that sounds very plausible.

Well, we have you and we have me - who else is on board?

Hate to say it, but you only have you so far.

Nope. Lots of places can't pay that money. They'll still try to interview and hire people but they can't compete.

There is a very wide range of salary, skill and experience in tech. A single figure is not of much use. If anything it might be ruined by the vast majority of cheap companies (i.e. outside of Google/Apple/Facebook/... )

Outside of lying, I've not seen an implementation of this that fully accounts for compensation outside of base salary (bonus, options, 401k matching, benefits, etc). I suspect some are entering only base, some base+bonus, or similar.

LinkedIn Salary appears to be reporting base salaries and total compensation.

Not necessarily lie. People, that earn less, are likely to be ashamed (even anonymously) of their salary, while big earners are prone to boast on every occasion.

Empirically, we see the opposite happening and compensation data on websites like Glassdoor is actually too low.

How do you know that the data on Glassdoor is too low? What is your source of truth? Anecdote? Glassdoor currently reports the average base salary for the job title "Senior Software Engineer" in SF Bay Area to be ~$130K which sounds about right to me, again, with nothing but anecdotal truth reference.

If you go by what people say on HN, then of course everyone in software makes $300K a year and has millions in stock options (and gets 5 job offers in their inbox a week).

I know how much I am paid, how much some of my coworkers are paid, how much my company pays new grads, and what the average salary on Glassdoor is for my company. The average is lower than what new grads got in 2016.

On top of the incorrect base salary, the information on bonuses/equity is extremely limited. The average that is listed is also incorrect.

to be fair, glassdoor still averages in salaries from 5 years ago into the current salary

That's part of it. People with higher salaries don't contribute for privacy reasons. They're also missing bonus/equity data for many firms.

Assuming you wanted to work for LinkedIn and Microsoft I'd definitely encourage you to lie about your current salary just in case they were so inclined to look it up and leverage it against you.

I don't know how widely it's used, but it seems like LinkedIn could some how use Work Number [1] to allow people to verify their salary.

[1] https://www.theworknumber.com/

Reply on news.ycombinator.com