Small sample size leads to greater variance, naturally. Better samples with no location:
The point is, changing programmer to engineer changes the rankings and salaries dramatically. pg called it noise.
don't assume that any links here withstood the submitter's professional scrutiny (are you serious?)
It wasn't a link; my understanding is the submitter (who happened to be you) specifically typed in all those parameters themselves. I apologize for coming off a bit cross; I almost always avoid that.
I'm not seeing this
I said the query I posted (with Engineer), not yours.
Anyway, it was a great idea, and anybody should be able to submit whatever they want and let the public judge; in this case, the post title seemed to overstate what it was showing. It was called "Salaries in SF (2009)", as if it was a report with analysis, but it was something linking to a search result, something likely to be different two weeks or two days from now. Maybe a blog post would have been better, as it would allow for multiple runs, analysis of the strengths and weaknesses of the methodology, and preserve the state of the results in the form of charts, as well.
Small sample size leads to greater variance, naturally. Better samples with no location:
The point is, changing programmer to engineer changes the rankings and salaries dramatically. pg called it noise.
don't assume that any links here withstood the submitter's professional scrutiny (are you serious?)
It wasn't a link; my understanding is the submitter (who happened to be you) specifically typed in all those parameters themselves. I apologize for coming off a bit cross; I almost always avoid that.
I'm not seeing this
I said the query I posted (with Engineer), not yours.
Anyway, it was a great idea, and anybody should be able to submit whatever they want and let the public judge; in this case, the post title seemed to overstate what it was showing. It was called "Salaries in SF (2009)", as if it was a report with analysis, but it was something linking to a search result, something likely to be different two weeks or two days from now. Maybe a blog post would have been better, as it would allow for multiple runs, analysis of the strengths and weaknesses of the methodology, and preserve the state of the results in the form of charts, as well.