More Charts from HN Survey (Income, Age, Work Hours, Work Exp.) | Hacker News Reader

More Charts from HN Survey (Income, Age, Work Hours, Work Exp.)

vonsharp.net

49 pointsviggity34 comments

The income histogram has zeroes cut off on the horizontal axis. Or is it just me?

Yes, starting at 105,000 and on up to 300,000 (the max on the chart) the last zero seems to have been cut off,

I don't think it's just the histogram. Check out the Age vs Income plot. There's no way people aren't making over 100k.

I cut off the age vs income plot at 100k because otherwise it would severely compress the points where it would be really hard to discern anything from it. There are people who reported over 100K.

How about reporting in 1000's, instead of including every zero? Typical in financial statements.

I wanted to do that, but I'm not proficient enough in excel to do it quickly. Surprisingly enough, even getting the histograms was kind of a pain in the ass. I spent an hour and a half getting those 4 charts done.

Yeah, its always shocked me that Microsoft couldn't be bothered to put in decent histogram fucntionality where you can choose your buckets and it'll put together a histogram.

I didn't even notice, I guess it is some sort of bug (or "feature") with excel, because the data driving that chart is correct. The numbers are in the correct order (increasing in 5K increments), but the second half has a zero clipped for some reason.

Perhaps decreasing the font (I use Numbers so don't know whether it is possible in Excel) of the horiz. axis will allow the full text to appear (if you can't display in 1,000's).

Staring at the Income/Frequency histogram, I am wondering why there is more clusering around the 10s and not the 5s. Ex: more salaries clustered around 60K, 70K, and 80K than around 65K, 75K, and 85K.

I wonder if there is a tendency (intended or otherwise) for employers/managers/etc to push salaries towards these numbers. Do they seem more 'round'?

I have some experience here, but I don't know if its really a trend. I was bumped up a while back from an 'uneven' salary to one of those clustering points. It was a small, strange, increase amount, but it was a welcome increase nonetheless!

I was surprised to see that the number of respondents working "40 hours a week" more than doubled any other entry. I listed 65 and felt sure I would be just average. FML lol

Most of the 40 hours a week people would probably be people just working full time jobs with no startup on the side.

Yeah, that's me right now. I'm taking a vacation from startups, so even though I still program all weekend, it's not "work" anymore. ;)

I wasn't sure how to respond to that question. I work 40 hours a week at my day job... but I work another 30+ hours a week on other projects (startup projects, etc).

What did you report in the survey? Some people probably reported just their day job and others their total work.

If anyone ever wants to do this again, they should try to see where interpretation was difficult and make it clear in the next survey.

I don't recall, but I think I just settled on 50 or 60, because I didn't want to skew results too incorrectly.

http://mathforum.org/kb/thread.jspa?threadID=194473&tsta...

Alright. Who's the 26yr. old with 900k income?

I'll guess a troll or Matt Mullenweg ;)

The age distribution is absolutely beautiful.

And kind of scary. It looks like a lot of people drop out of programming after 30...

Of course, the other possibility (probability) is that most HackerNews readers are under 30.

Seem to be a surprisingly (?) high number of 300k+ earners. All Silicon Valley?

Let me know if you're interested in seeing more visualizations.

Other possible interesting graphs:

  * Family type v. income
  * Number of years in industry v. income
  * Hours per week v. family status
  * Age v. Employment type
  * Level of education v. Employment type
  * Marital status v. Employment type
  * Age v. Hours per week

Age vs. Income as a density plot... the points are too close together to be meaningful.

If you have any visualisation techniques that are surprising or especially revealing I would be very interested both to look at the data and to learn about data visualisation.

Thanks for doing that.

Maybe also: Income vs Education. Income vs Location (maybe aggregated by country, state or city). And basically every other possible combination :)

Does there exist any sort of web app that can take rows of data and allow the user to run statistical reports on it?

Outstanding work! I'm glad to see other people making something interesting out of the data gathered.

A bunch of young, inexperienced single men making less than 50k?

Who would have thought.

That would be less than 500k if I'm reading this correctly (although a significant portion < 100k, to be sure).

Where do you get "less than 50k"?

I found it interesting that only 1.8% of the population seems to be female

Me too. Seems to be increasing. :-P

Reply on news.ycombinator.com