Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Poll: HN readers, where's your residence?

news.ycombinator.com

Post reply on HN
Print view View original

361–370 of 506 posts

Re: Poll: HN readers, where's your residence?

#361

Post by skoob »

zerostar07 wrote: post #322
tokenadult wrote: post #294

Even if the poll is posted for many hours on HN's front page, we shouldn't rely on it for an inference of where HN participants live, in the level of detail shown in the poll. Just three days ago HN had a good thread about gaps in statistics knowledge even among persons with a university education, http://news.ycombinator.com/item?id=3285742 and polls like this on HN are a good illustration (in the comments) about wh…

Nobody argues that this is not scientific research, but that doesn't mean it's not interesting or useful. It also violates some of your expectations: the majority is not in the US, India is underrepresented despite English being their second language. And if i might add, despite their national focus on entrepreneurship, the UK numbers are low.

As I'm writing this the numbers are 2186 for the US vs 496 for the UK. The US population is almost five times the size of the UK, so the UK is actually overrepresented compared to the US.

Re: Poll: HN readers, where's your residence?

#365

Post by ruycer »

altuzar wrote: post #360
armandososa wrote: post #333

Uhm, I'm in Mexico and Mexico is in North America. http://en.wikipedia.org/wiki/North_America http://en.wikipedia.org/wiki/Central_America http://en.wikipedia.org/wiki/South_America See?

Je suis @ Mexico aussi. We are North America!!!

I am in Mexico too. So either North America if you go for a geographical division, or Latin America, if you wish to characterize culturally.

Re: Poll: HN readers, where's your residence?

#368

Post by drewcrawford »

j_baker wrote: post #355
tokenadult wrote: post #294

Even if the poll is posted for many hours on HN's front page, we shouldn't rely on it for an inference of where HN participants live, in the level of detail shown in the poll. Just three days ago HN had a good thread about gaps in statistics knowledge even among persons with a university education, http://news.ycombinator.com/item?id=3285742 and polls like this on HN are a good illustration (in the comments) about wh…

Forgive me if this is a dumb question, but what is "voluntary response data"? Is the author arguing that the only statistically valid way to do a study is to put a gun to peoples' heads and demand that they take a survey? I can't think of many ethically sound ways to conduct a study without responses being voluntary.

Think about if the poll were (the horror!) political. The person who responds to that kind of poll is the intersection of:

1) An HN reader

2) Somebody who thinks politics on HN is a good idea

3) Somebody who is on HN / has free time TODAY, skewing the distribution in favor of certain verticals that are in a low point of their seasonal cycle or against startups in a coding marathon

4) Somebody who personally cares about politics sufficiently to have an opinion strong enough to want to be counted, this is likely to skew the distribution in favor of extremes

In contrast, if pg picks 1000 random HN users and personally asks them the poll question via e-mail, some of these biases evaporate. And you lose the "interested bystander" effect whereby more interested people tend to want to take the poll, because the only way to take it is by personal invitation.

Re: Poll: HN readers, where's your residence?

#369

Post by tonio09 »

postfuturist wrote: post #352
armandososa wrote: post #333

Uhm, I'm in Mexico and Mexico is in North America. http://en.wikipedia.org/wiki/North_America http://en.wikipedia.org/wiki/Central_America http://en.wikipedia.org/wiki/South_America See?

My first thought when I read the poll was, "What about everything between the US and South America?". I think the way these polls are written tells more about the person writing the poll than anything else. I can't imagine the usefulness of splitting the US into NYC/SF and everything else. Likewise with splitting the UK and the rest of Europe.

why not split US and EU? Clearly a statistically significant number of visitors are from NY/SF and UK.
Post reply on HN