Personality and Popularity in the Speakeasy (A Study!)

[Psyche]

so i reverse scored every unpopular trait
to reverse score a trait is just to compute 1-thatscore
my measure of depression becomes a measure of contentedness, and so forth

and i've found all sorts of neat things by reanalyzing things

for example, only about 10 of the 50+ traits are overrepresented in the community (compared to about half before); not even one of those overrepresented traits

significantly

correlates with popularity
all of this just confirms again that the rare traits in community are the popular ones

These 18 traits are significantly with popularity (! before a trait just indicates that it's been reverse-scored; read it as "not <Trait>"):
[1] Trust Artistic.interests Cooperation Conscientiousness ! Depression
[6] Morality Achievement.striving Cautiousness Self.discipline ! Anger
[11] Modesty ! Excitement.seeking ! Assertiveness Dutifulness Self.expression
[16] Self.efficacy ! Neuroticism Adventurousness

trait underrepresentation in the community still correlated with trait popularity (though more weakly; r = -.55, p = 1.837e-05)...

as expected i haven't found anything contrary to what i already know here
but this reverse scoring should help a bit grappling with Mina's question

also there will be a spreadsheet in the public folder before you know it that tries to assess how likeable everyone is...

[Psyche]

made the likeability spreadsheet
it's a mean of individuals' traits weighted according to how much each correlates with popularity (and unpopular traits reverse-scored)

tbh i'm not sure what any of it means
a lot of highly ranked people were found to have low likeability by my and IBM's measure
while a lot of lowly ranked people were found to have the opposite
overall the likeability score correlates positively with popularity (r = .28, p = 0.0006165)

that's moderate; it's not nothing; it would make a nomoral psychologist proud
but it's also not the .7 (or maybe just .5?) i found between trait underrepresentation and trait popularity
why, if i calculated likeability using each individual's Trust alone, I'd have a better prediction of popularity (r > .3)

so either having some popular traits makes one likely to have unpopular traits or something, or i'm just not putting the information together in the right way
i'll want to try some more formal methods to handle this problem

in the mean time i really should get to finishing mina's request

[Majiffy]

I'm pretty high with all those negative traits explan my ranking in SUPP mr smartypants bookman


Yeah took your little study and shoved it didn't i

[Psyche]

a friend suggested the relationship between prevalence in the community and popularity as far as traits go is an artifact of
1) that certain traits like Depression are "objectively worse" than others like Morality
2) internet forums like ours are cesspools of bad traits

i tried to apply reverse-scoring to this issue
Depression might seem "objectively bad", so ~Depression (the reverse scoring of that trait) probably seems great

this just enables a reorganization of her argument

originally, the idea was more two-dimensional than i let on

my friend's argument was that what really makes a trait popular is how "good" it is;
conscientiousness is good and depression is bad, for example, according to this logic
if one accepts this hypothesis, then my observation that popular traits are underrepresented in our community is the same as the observation that our community is a bad one - rich in bad traits and lacking in good traits
scary stuff

by reverse-scoring bad traits i take badness out of the equation in a sense
but even when i do this i observe that "good" traits are underrepresented in our community and the cesspool hypothesis continues to live

it's even more obvious now than before that i won't be able to crack this puzzle without performing this same study in other communities

[SleepyKrew]

is there a way to see most likable to least likable

[Psyche]

not until it's better



this combines my composite personality profile graph with my trait prevalence vs trait popularity graph
instead of ordering traits by average score, i ordered by correlation with probability
and finally i overlayed a line of best fit relating average score with the rank of a trait's popularity correlation

overrepresented traits generally occur near the bottom (least popular)
and underrepresented traits generally occur near the top (most popular)

mina's, my friend's, and my hypotheses all seem to explain this relationship and can't really be adjudicated without further research in other communities
besides sharing the graphs she requested, i'm probably done thinking about this issue for now

i'm going to focus on prediction and producing better likeability scores from here out

[Psyche]

multiple regression is extremely easu using r i really should have done it near the very beginning of my data analysis
i think in general out of laziness of a lack of imagination i've stuck too closely to basic linear analyses when there's a lot more really cool statistics i can try out

[Kublai Khan]

Oh you fucking killjoy

Are you seriously trying to get this shut down because a string of 6000 words likely with just a username and (rarely) first name attached goes to a source which probably won't but may read it even though they likely have no clue who the user is and can't really do fuck all about it

This is just fucking ridiculous. Come on.

Yes, it's literally my job to do exactly that. Some people actually enjoy having a modicum of confidentiality and object to having it violated on the basis of hopeful indifference by a corporation whose terms of service literally state that they are looking to collect as much information as possible for marketing purposes.

In the meantime I've suggested a workaround that doesn't violate site policy. You're welcome.

[Majiffy]

it's even more obvious now than before that i won't be able to crack this puzzle without performing this same study in other communities

Lol do it at EM

[Oman]

Some people actually enjoy having a modicum of confidentiality

These people are fooling themselves and I think the site would do better not to indulge the fantasy.

[Psyche]

deasvail was robbed

[Psyche]

made the likeability spreadsheet
it's a mean of individuals' traits weighted according to how much each correlates with popularity (and unpopular traits reverse-scored)

tbh i'm not sure what any of it means
a lot of highly ranked people were found to have low likeability by my and IBM's measure
while a lot of lowly ranked people were found to have the opposite
overall the likeability score correlates positively with popularity (r = .28, p = 0.0006165)

that's moderate; it's not nothing; it would make a nomoral psychologist proud
but it's also not the .7 (or maybe just .5?) i found between trait underrepresentation and trait popularity
why, if i calculated likeability using each individual's Trust alone, I'd have a better prediction of popularity (r > .3)

so either having some popular traits makes one likely to have unpopular traits or something, or i'm just not putting the information together in the right way
i'll want to try some more formal methods to handle this problem

using multiple regression i was able to make it to .5, which is pretty respectable!
at some point i'll see if these results can predict the results of other popularity contests with sufficient reliability. And of course there's future contests, too.

But, uh, I'm starting to feel fatigue about all this and so may take a day or two's break from this project.

[Bellaphant]

Can I /in and consent to whatever? I want a profile:) when you are less fatigued


C

[Ether]

I made a version of the (currently) ranked data that's easier to read.

Also jesus I am at the crazy extremes on this thing. Some of it I can totally buy, but, uh. I really,

really

hope I'm not the least depressed person here.

[Davsto]

I made a version of the (currently) ranked data that's easier to read.

Also jesus I am at the crazy extremes on this thing. Some of it I can totally buy, but, uh. I really,

really

hope I'm not the least depressed person here.

This is more interesting to me

For example, I am apparently ranked 145 for anger, but it is still one of my more present traits because apparently everyone on this site is angry (who knew)

Also me being ranked 5 for adventurousness is frankly almost comical

[Ether]

I am THE MOST ADVENTUROUS PERSON ON THE SITE.

(No I'm not.)

[Iecerint]

/in

[Iecerint]

multiple regression is extremely easu using r i really should have done it near the very beginning of my data analysis
i think in general out of laziness of a lack of imagination i've stuck too closely to basic linear analyses when there's a lot more really cool statistics i can try out

Try doing polynomial regression on a few indicators maybe.

All you have to do is center the variable, then add a regressor that's the square (or cube, etc., but it gets hard to interpret after cubes) of the centered indicator. The polynomial is important if the squared regressor is significant.

With personality indicators for some variables there's a "sweet spot" that's ideal whereas the extremes are pathologic. If that's true about popularity, then at least testing a polynomial regression model is probably really important for model specification.

[Iecerint]

I'd really like the SE 6000 word thing you wrote (or just the words that it grabs for me). I wanted to do the IBM thing before, but I don't really use Twitter or anything, so it's hard to get 6000 words that aren't theses.

[MattP]

Do high scores mean you're not the trait listed? So high scores for positive traits is bad but high scores for negative traits is good?

[Ether]

For the rankings comparing you to other 'Scummers, yes. #1 would be the person who has the most of the trait; #147 the least.

For your personal profile, higher is more.

[Papa Zito]

gdi Psyche you could've told me I was already in this

[PJ.]

I'm 2nd in hedonism. Apparently depression is one of my highest traits but I'm rated 99th? My highest traits were a combination of hedonism, anger, love, depression, self enhancement and self conciousness. I'm apparently Cersei Lannister.

[Iecerint]

Speakeasy has a thread whose sole purpose is as a dumping ground for angst, so it makes sense that depression would be overrepresented.

[PJ.]

Speakeasy has a thread whose sole purpose is as a dumping ground for angst, so it makes sense that depression would be overrepresented.

I'd argue that's the sole purpose of the entire speakeasy.