What’s the point of the Chromosomes question? Once you know someone is a cis male or a trans female, does knowing that they have a Y chromosome tell you that much more?
The “White (Hispanic)”—“White (non-Hispanic)” dichotomy is weird to non-Americans, and you may want to add an “Other” answer—or what are (say) Arabs or Maori supposed to answer?
I’d split the Children question into “how many children you have” (write-in) and “do you hope to have more children in the future” (with answers “yes, soon”, “yes, later on in my life”, “no” and “not sure”).
If you don’t want to add more options to “Political” (e.g. libertarian socialist), please add a “None of the above” answer. (Also, I’d say “most strongly identify or lean towards”.) BTW “liberal” and “libertarian” have other meanings (especially outside America), but that’s not a big deal given you give examples.
As IIRC was suggested after the last survey, you might link to a non-amateur Internet IQ test, if such a thing exists.
Maybe ask both total karma and last-30-day karma?
What counts as “intelligent life” in P(Aliens)? I’m assuming that octopus/crow/dolphin/bonobo-level intelligence doesn’t count, and that humans right after the Upper Palaeolithic Revolution do.
In P(Cryonics) you might want to replace “average” with “randomly chosen”, if you mean the average of the probabilities.
“Singularity” is ambiguous—do you mean P(singularity before the year X|singularity ever) = P(singularity after the year X|singularity ever), or P(singularity before the year X) = P(singularity after the year X or never)?
I’d split the Children question into “how many children you have” (write-in) and “do you hope to have more children in the future” (with answers “yes, soon”, “yes, later on in my life”, “no” and “not sure”).
Yes. That question is asking two different things, which a survey question should never do (in the extreme, that’s called a “loaded question”).
Well, if one interprets “asking two different things” broadly, any question with more than two answers is doing that, e.g. the one about sexual orientation is asking whether you’re sexually attracted to males and whether you’re sexually attracted to females. (Maybe I’d split that one, too, as Facebook does.)
It depends on whether you end up covering all of the possibilities. It’s fine to ask something like this (though it’s phrased badly and confusing):
Which of the following movies have you seen?
Iron Man
The Hulk
Both
Neither
But this would not be good:
Have you seen Iron Man?
Yes, and I liked it
No, I hate comic book movies
The point is to not have any respondents who could not truthfully select one of the options in a single-punch list, or at least one option in a multi-punch list. The above question offers no response for folks who haven’t seen Iron Man but don’t hate comic book movies, or folks who have seen Iron Man but didn’t like it.
Well, technically the question about children in the draft doesn’t do that (except for ambiguities such as how to count dead children, or people who are exactly indifferent about having children in the future), but I still think it divides personspace in a weird way.
Yes, technically there’s no excluded middle problem there. Instead, it’s gathering information about future child-wanting only for those with no children, which is a different problem.
In “Religious Views”, I’d have “None of the above”, for apatheism (not giving a damn about whether any deity exists), ignosticism (not agreeing that the question even makes sense in the first place), etc. Also, what is meant by “spiritual”?
Actually, I remember proposing to ask height in the next survey about half a year ago, but I no longer remember why for sure (something about height-IQ correlations?) nor whether I was entirely serious...
It’ll sure be obnoxious if we genetic engineer ourselves in to increasingly taller and shorter-lived bodies. Seems like a classic prisoner’s dilemna type scenario, similar to steroid abuse.
Biological engineering will enable us to remake ourselves in the image of our dreams—the problem is that we’re kind of stupid. My impression is that if we’re starting from something like current culture, the default will be to try to create children who are tall, lean, hypomanic, good at taking standardized tests, and probably blond.
I assume they will. Even if invulnerability to fashion has a genetic basis, I doubt that very many people would select for it. I take that back—if it’s possible to select for stable imprinting on “tradition” (what you’ve grown up with), some parents will want that.
It would be interesting to see whether height correlates with other answers.
It correlates with income and ability to become president of the US. It could correlate with some P() questions. If it does that would be interesting to know.
Chromosomes makes that info easier to process and is useful in case a bunch of people put their gender as “other” or don’t understand the gender question.
Yep. Unless you have had your DNA sequenced or the like, you don’t know your chromosomes; there are a number of unusual genotypes that are not obvious.
Besides which, hormones matter a whole lot more for human sexual differentiation than chromosomes. Birds are different, and a lot more like the naive idea of “chromosomes > sex characteristics” (which is why you sometimes get bilaterally gynandromorphic birds when a pair of zygotes—one male, one female—fuse in the egg).
Good point- without DNA sequencing, we’re guessing about specific genes based only on their expression, when their expression can be muted by other factors.
Not that low. (Unless he has children, at least.) EDIT: I’m pretty sure I read a longer version of that article, also mentioning Olympic sex tests etc., but I can’t find it anymore.
In any event, that reminds me that having more than two X chromosomes, or more than one Y chromosome, doesn’t matter much, so if the question is kept I’d specify that XY also includes XYY, XYYY etc. and XX also includes XXX, XXXX, etc.
What army1987 said. Naturally intersexed folks are about 1% of the population, though that number probably includes some non-chromosonal differences and excludes some chromosonal differences.
And I said “over 95%” because I knew it would be at least 95% if I thought about it, but I hadn’t yet.
IIRC, a suggestion I saw and I liked was to ask “What sex were you assigned at birth?” (Male/Female/Other) and “What gender do you currently identify as?” (Male/Female/Other).
It also creates potential time cost for people looking up what XX and XY chromosomes refer to. If you leave this question in the survey, can you at least include a heuristic for the uninformed, such as “heuristic: biologically female ⇒ XX; biologically male ⇒ XY)”?
What’s the point of the Chromosomes question? Once you know someone is a cis male or a trans female, does knowing that they have a Y chromosome tell you that much more?
The “White (Hispanic)”—“White (non-Hispanic)” dichotomy is weird to non-Americans, and you may want to add an “Other” answer—or what are (say) Arabs or Maori supposed to answer?
I’d split the Children question into “how many children you have” (write-in) and “do you hope to have more children in the future” (with answers “yes, soon”, “yes, later on in my life”, “no” and “not sure”).
If you don’t want to add more options to “Political” (e.g. libertarian socialist), please add a “None of the above” answer. (Also, I’d say “most strongly identify or lean towards”.) BTW “liberal” and “libertarian” have other meanings (especially outside America), but that’s not a big deal given you give examples.
As IIRC was suggested after the last survey, you might link to a non-amateur Internet IQ test, if such a thing exists.
Maybe ask both total karma and last-30-day karma?
What counts as “intelligent life” in P(Aliens)? I’m assuming that octopus/crow/dolphin/bonobo-level intelligence doesn’t count, and that humans right after the Upper Palaeolithic Revolution do.
In P(Cryonics) you might want to replace “average” with “randomly chosen”, if you mean the average of the probabilities.
“Singularity” is ambiguous—do you mean P(singularity before the year X|singularity ever) = P(singularity after the year X|singularity ever), or P(singularity before the year X) = P(singularity after the year X or never)?
Yes. That question is asking two different things, which a survey question should never do (in the extreme, that’s called a “loaded question”).
Well, if one interprets “asking two different things” broadly, any question with more than two answers is doing that, e.g. the one about sexual orientation is asking whether you’re sexually attracted to males and whether you’re sexually attracted to females. (Maybe I’d split that one, too, as Facebook does.)
It depends on whether you end up covering all of the possibilities. It’s fine to ask something like this (though it’s phrased badly and confusing):
Iron Man
The Hulk
Both
Neither
But this would not be good:
Yes, and I liked it
No, I hate comic book movies
The point is to not have any respondents who could not truthfully select one of the options in a single-punch list, or at least one option in a multi-punch list. The above question offers no response for folks who haven’t seen Iron Man but don’t hate comic book movies, or folks who have seen Iron Man but didn’t like it.
Well, technically the question about children in the draft doesn’t do that (except for ambiguities such as how to count dead children, or people who are exactly indifferent about having children in the future), but I still think it divides personspace in a weird way.
Yes, technically there’s no excluded middle problem there. Instead, it’s gathering information about future child-wanting only for those with no children, which is a different problem.
In “Religious Views”, I’d have “None of the above”, for apatheism (not giving a damn about whether any deity exists), ignosticism (not agreeing that the question even makes sense in the first place), etc. Also, what is meant by “spiritual”?
Other questions I’d like to see:
native language;
height;
income
Why height?
Actually, I remember proposing to ask height in the next survey about half a year ago, but I no longer remember why for sure (something about height-IQ correlations?) nor whether I was entirely serious...
Off topic, but there’s some evidence that shorter people live longer:
http://www.sciencedirect.com/science/article/pii/S0024320502025031
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1071721/
It’ll sure be obnoxious if we genetic engineer ourselves in to increasingly taller and shorter-lived bodies. Seems like a classic prisoner’s dilemna type scenario, similar to steroid abuse.
Ohhh, yeah.
Biological engineering will enable us to remake ourselves in the image of our dreams—the problem is that we’re kind of stupid. My impression is that if we’re starting from something like current culture, the default will be to try to create children who are tall, lean, hypomanic, good at taking standardized tests, and probably blond.
Hm. I wonder if designer baby characteristics will go in and out of style the same way baby names do.
I assume they will. Even if invulnerability to fashion has a genetic basis, I doubt that very many people would select for it. I take that back—if it’s possible to select for stable imprinting on “tradition” (what you’ve grown up with), some parents will want that.
It would be interesting to see whether height correlates with other answers.
It correlates with income and ability to become president of the US. It could correlate with some P() questions. If it does that would be interesting to know.
The correlation for presidents is weaker than I thought.
Chromosomes makes that info easier to process and is useful in case a bunch of people put their gender as “other” or don’t understand the gender question.
I don’t know the answer to the chromosomes question. I could guess, and I would put over 95% on it, but it still seems weird.
Yep. Unless you have had your DNA sequenced or the like, you don’t know your chromosomes; there are a number of unusual genotypes that are not obvious.
Besides which, hormones matter a whole lot more for human sexual differentiation than chromosomes. Birds are different, and a lot more like the naive idea of “chromosomes > sex characteristics” (which is why you sometimes get bilaterally gynandromorphic birds when a pair of zygotes—one male, one female—fuse in the egg).
Good point- without DNA sequencing, we’re guessing about specific genes based only on their expression, when their expression can be muted by other factors.
that seems low.
Not that low. (Unless he has children, at least.) EDIT: I’m pretty sure I read a longer version of that article, also mentioning Olympic sex tests etc., but I can’t find it anymore.
In any event, that reminds me that having more than two X chromosomes, or more than one Y chromosome, doesn’t matter much, so if the question is kept I’d specify that XY also includes XYY, XYYY etc. and XX also includes XXX, XXXX, etc.
What army1987 said. Naturally intersexed folks are about 1% of the population, though that number probably includes some non-chromosonal differences and excludes some chromosonal differences.
And I said “over 95%” because I knew it would be at least 95% if I thought about it, but I hadn’t yet.
IIRC, a suggestion I saw and I liked was to ask “What sex were you assigned at birth?” (Male/Female/Other) and “What gender do you currently identify as?” (Male/Female/Other).
I support this. No need to bring chromosomes into this.
It also creates potential time cost for people looking up what XX and XY chromosomes refer to. If you leave this question in the survey, can you at least include a heuristic for the uninformed, such as “heuristic: biologically female ⇒ XX; biologically male ⇒ XY)”?