I was thinking about the fact that there is probably a difference between active LWers versus lurkers or newbies. So I looked at the data for the Top 100 users (actually Top 107, because there was a tie). This happily coincided with the nice Schelling point of 1000 karma. (make sense, because people are likely to round to that number.) To me, this reads as “has been actively posting for at least a year”.
So, some data on 1000+ karma people:
Slightly more likely to be male: 92.5% Male, 7.4% Female
NOTE: First percentage is for 1000+ users, second number is for all survey respondents
Much more likely to be polyamorous:
Prefer mono 36% v. 54% Prefer poly 24% v. 13% Uncertain 33% v. 30% Other 4% v. 2%
About the same Age: average 28.6 v. 27.8
About as likely to be single 51% v. 53%
Equally likely to be vegetarian 12%
Much more likely to use modafinil at least once per month: 15% v. 4%
About equal on intelligence tests
SAT out of 1600: 1509 v. 1486 SAT out of 2400: 2260 v. 2319 Self reported IQ: 138.5 v. 138.7 online IQ test: 127 v. 126 ACT score: 33.3 v. 32.7
Similar income 50k
Slightly lower Autism quotient: average 22 v. 24
More likely to choose torture Torture: 42% v. 22% Dust Specks: 29% v. 37%
More likely to cooperate in a Prisoner’s Dilemma: Cooperate: 36% v. 27% Defect: 20% v. 29%
Some notes: Yes, I realize my data analysis methods are not the best. Namely, that instead of comparing the people with >1000 karma to the people with <100 karma, which would have been more accurate, I just compared them to the overall results (which includes their answers). I did this because it takes much less time.
Also, a hint for other people playing with the data in Excel format: A lot of the numbers are in text format, and a pain to convert to numeric format in a way that allows you to manipulate them. The easiest work around (so long as you don’t want to do anything complicated) is to just paste the needed columns either into google spreadsheet, or into another Excel sheet that’s been formatted numerically. If you want to do something complicated you probably need to find the “right” way to fix it.
Also, a hint for other people playing with the data in Excel format: A lot of the numbers are in text format, and a pain to convert to numeric format in a way that allows you to manipulate them. The easiest work around (so long as you don’t want to do anything complicated) is to just paste the needed columns either into google spreadsheet, or into another Excel sheet that’s been formatted numerically. If you want to do something complicated you probably need to find the “right” way to fix it.
multiplying text by 1 or adding zero can often force auto conversion in excel. You can do this by past as values multiply. Shortcut keys are copy 1 then highlight data ALT+e s then v m enter.
I had prepared the following post before I came across the one by daenerys.
Here are the statistics of the members who claim to have 4000 karma or more. The sample was too small and I was too lazy to fix the data so I used medians (I did it manually). yvain can definitely do a better job since he has the data already fixed and can access the unpublished data.
0 INFPs with over 4k? Well, it looks like that has outed me as not filling in this year’s survey! Well, unless I was the type to be squeamish about revealing karma or identifying information in such a case (not likely!)
Top 100 Users’ Data, aka Karma 1000+
I was thinking about the fact that there is probably a difference between active LWers versus lurkers or newbies. So I looked at the data for the Top 100 users (actually Top 107, because there was a tie). This happily coincided with the nice Schelling point of 1000 karma. (make sense, because people are likely to round to that number.) To me, this reads as “has been actively posting for at least a year”.
So, some data on 1000+ karma people:
Slightly more likely to be male:
92.5% Male, 7.4% Female
NOTE: First percentage is for 1000+ users, second number is for all survey respondents
Much more likely to be polyamorous:
Prefer mono 36% v. 54%
Prefer poly 24% v. 13%
Uncertain 33% v. 30%
Other 4% v. 2%
About the same Age:
average 28.6 v. 27.8
About as likely to be single
51% v. 53%
Equally likely to be vegetarian
12%
Much more likely to use modafinil at least once per month:
15% v. 4%
About equal on intelligence tests
SAT out of 1600: 1509 v. 1486
SAT out of 2400: 2260 v. 2319
Self reported IQ: 138.5 v. 138.7
online IQ test: 127 v. 126
ACT score: 33.3 v. 32.7
Similar income
50k
Slightly lower Autism quotient:
average 22 v. 24
More likely to choose torture
Torture: 42% v. 22%
Dust Specks: 29% v. 37%
More likely to cooperate in a Prisoner’s Dilemma:
Cooperate: 36% v. 27%
Defect: 20% v. 29%
Some notes: Yes, I realize my data analysis methods are not the best. Namely, that instead of comparing the people with >1000 karma to the people with <100 karma, which would have been more accurate, I just compared them to the overall results (which includes their answers). I did this because it takes much less time.
Also, a hint for other people playing with the data in Excel format: A lot of the numbers are in text format, and a pain to convert to numeric format in a way that allows you to manipulate them. The easiest work around (so long as you don’t want to do anything complicated) is to just paste the needed columns either into google spreadsheet, or into another Excel sheet that’s been formatted numerically. If you want to do something complicated you probably need to find the “right” way to fix it.
multiplying text by 1 or adding zero can often force auto conversion in excel. You can do this by past as values multiply. Shortcut keys are copy 1 then highlight data ALT+e s then v m enter.
I had prepared the following post before I came across the one by daenerys.
Here are the statistics of the members who claim to have 4000 karma or more. The sample was too small and I was too lazy to fix the data so I used medians (I did it manually). yvain can definitely do a better job since he has the data already fixed and can access the unpublished data.
Probabilities:
PManyWorld: 67.5%
PAliens: 80%
PAliens2: 20%
Psupernatural: 0.1%
PGod: 0.03%
PReligion: 0.00000005
PCryonics: 8.5
PAntiagathics: 20%
PSimulation: 10%
PWarming: 80%
PGlobalcatastrophicrisk: 75%
Singulartity: 2070
TypeofGlobalCatastrophicRisk: 10 Unfriendly AI, 7 Pandemic bioengineered, 3 Nanotech / grey goo, 2 Nuclear war, 1 Unknown Unknowns, 1 unsure
Personality:
MyersBriggs: 5 INTJ, 2 INTP, 2 ENFP, 1 ENTJ, 1 ISTP
BigFiveO: 80
BigFiveC: 35
BigFiveE: 37
BigFiveA: 38
BigFiveN: 37
IQTest: 135
AutismScore: 23
Politics:
5 Socialist, 12 Liberal, 3 Conservative, 6 Libertarian
AlternativeAlternativePolitics: 3 Moldbuggian, 2 Futarchist, 1 Technocratic, 1 Pragmatist (the rest were unremarkable).
PoliticalCompassLeftRight: 1.25
PoliticalCompassLiberty: −5.28
Vegetarians: 16%
SRS: 36%
0 INFPs with over 4k? Well, it looks like that has outed me as not filling in this year’s survey! Well, unless I was the type to be squeamish about revealing karma or identifying information in such a case (not likely!)