Working in survey based stats all day, I can tell you that the easiest junk answers to filter out are the ones you're seeing at the moment. The plausible looking results you used to get aren't necessarily as good as you think they were.
Working in survey based stats all day, I can tell you that the easiest junk answers to filter out are the ones you're seeing at the moment. The plausible looking results you used to get aren't necessarily as good as you think they were.
What I find more interesting is that this thread has been allowed to survive rather than being downmodded. And that tells me that some influential HN'ers consider the responses to this survey meaningful - either individually or in aggregate. One of the often ignored aspects of HN is that to a meaningful degree it is an extension of people's workplaces. I find it handy to remember that when participating.
I rather doubt that the point of the survey is actually collecting data on HN'ers ages. Detailed demographic information about people on the internet is readily obtainable from other sources.
On the other hand, information about who is likely to "vandalize" HN is more difficult to collect and highly relevant, and this is the jist of PG's response as I read it. I couldn't really come up with a better term for crapping on a poll than "vandalize." The scare quotes are there because what is happening is normal internet behavior.
Perhaps the vandalism is somewhat invited by the absurdity of the youngest age and the quasi-absurdity of the oldest - does it really matter from an HN demographic standpoint to distinguish between those over 80 and those over 90 - both cohorts must have came to computing as adults.
Not that this excuses the behavior, and perhaps the invitation to vandalism is intentional because the wording and presentation of many attempts at data collection via ASK HN is awkward enough to invite meanness of which vandalism is one form.
[EDIT] Indeed, the creator of the poll claims to be studying quantitative psychology.
Being unwilling to share personal information is no reason to vandalize it.
Wanting to "click all the buttons" is only a valid excuse if you are under the age of 8.
What's silly is defending the dickheads who purposely ruin the poll.
Was phreaking illegal? Totally. Was it unethical? Yes. Do we not consider it a part of hacker history and culture? You can disagree with people and say what they did was not morally sound, but it's silly to say people don't belong on a hacker site because they have an anarchist kind of worldview.
This isn't nearly as interesting of a discussion as you think it is.
I'm glad we found someone who can objectively decide what's interesting! Now you can suggest books and movies to me! Do you work for Netflix?
But seriously, I never made a value judgement about the poll. The poll costs real time, and real money to host. But data from a poll is not a window. Adding noise into a poll is not breaking a window.
This blatant inaccuracy is actually harmless, because it basically invalidates the two buckets which were already going to be an insignificant part of the poll. Absent any response I would've said those buckets are there to absorb troublemakers, but pg's response implies I'm giving him too much credit. The bigger risk is people who are fudging by one or two buckets, giving plausible but inaccurate data. Hell, the whole poll could actually be inaccurate, but we're focusing on the fact that there are a few joke responses which can be discarded safely.
I get to decide this isn't interesting because I'm half of the discussion. And I'm telling you, this is boring.
I think that is too wide a definition of unacceptable.
While sure, this might be the case, it still doesn't negate the fact that fundamentally, HN's arrows (of all kinds) should be able to be "undone". Accidents happen, people are curious. We shouldn't blame the users for a system that is broken and relies on them to ignore long-standing standard UX practices when it comes to these types of elements. This poll could have only been abused so far if users were only allowed to pick one, even if they chose to pick the wrong one. Given how long people have been complaining about this, it doesn't surprise me they take advantage of it for the sake of proving a point.
In an ideal world, we should not even be allowed to see the results before voting, since it could bias us (it should not in this kind of poll, of course).
But I definitely agree the responses should not be visible until you've voted.