Gay StarCraft Players - Page 206

On January 12 2012 12:09 matjlav wrote:


Hm, I think you need to fix your understanding of statistics. No statistical test on that poll can show that the distributions are equal; it might just not prove that they're not equal. That's why the language is "reject" or "do not reject" the null hypothesis that the distributions are equal instead of "reject" and "accept." When you get a value of alpha above the threshhold probability, it's just showing you that if the distributions are equal, the given result is not all that unlikely, which doesn't constitute proof at all that they are in fact equal; it just leaves the possibility open, unlike if you get a very small alpha.

Also, I believe that the chi-squared test is appropriate for this sort of thing, not a simple t-test

Sorry, I just had to...

On January 12 2012 12:09 matjlav wrote:


Hm, I think you need to fix your understanding of statistics. No statistical test on that poll can show that the distributions are equal; it might just not prove that they're not equal. That's why the language is "reject" or "do not reject" the null hypothesis that the distributions are equal instead of "reject" and "accept." When you get a value of alpha above the threshhold probability, it's just showing you that if the distributions are equal, the given result is not all that unlikely, which doesn't constitute proof at all that they are in fact equal; it just leaves the possibility open, unlike if you get a very small alpha.

Also, I believe that the chi-squared test is appropriate for this sort of thing, not a simple t-test

Sorry, I just had to...

On January 12 2012 13:16 Klondikebar wrote:


Eh, you're right. It's been 4 years since I've dealt with that stuff so I'm a little rusty. I took the class freshman year and after that you do EVERYTHING in Stata and you never bother with simple stuff. And it doesn't help that the job that hired me to do statistics really has no idea what statistics is...all they want is connect the dot graphs.

Hopefully I'll pick it back up when I start studying for my CFA exams in a couple of days :/

I was always better at Economic theory than I was at statistics but part of getting the degree means taking Statistics. I really kinda loathe math. The only thing about it I like is that, when done correctly, it's always certain.

On January 12 2012 12:33 Troxle wrote:
Statistics....so glad I only have to deal with Calculus. That looks too confusing for my brain. (I attempted to read it, but I got lost in all the terminology and lack of understanding...)

On January 12 2012 22:21 adrenaLinG wrote:


Most advanced economics require substantial math though. Even at the undergraduate level econometrics and the intermediate courses have a good amount of it.

Also don't go for the CFA unless you have a job that requires it (portfolio manager or equity research), the designation is quite saturated atm

On January 12 2012 22:31 Roe wrote:

statistics harder than calculus? whaaaaaat?

On January 12 2012 23:30 Troxle wrote:


I've never taken statistics so I wouldn't know to be honest, just trying to understand what is being talked about right now is lost on me. If it was Calculus talk y'all would probably be the same way as I feel now though!

On January 12 2012 23:56 Klondikebar wrote:


Don't try to understand my posts...I heinously misstated a conclusion and used the wrong test. Although on the bright side I did go back and re-learn some stuff so I guess there's a silver lining. It's like trying to learn to walk again though. The stuff is so simple and stupid there really is no excuse for forgetting it other than my brain is just rotting.

On January 13 2012 04:49 Troxle wrote:


I don't understand all these tests you talk about haha! I'm used to my theorms and derivatives

On January 12 2012 23:30 Troxle wrote:


I've never taken statistics so I wouldn't know to be honest, just trying to understand what is being talked about right now is lost on me. If it was Calculus talk y'all would probably be the same way as I feel now though!

On January 13 2012 09:28 matjlav wrote:


Yeah, basic stats is much easier than calc. I've heard that advanced stats is a good deal harder than the basic stuff, though.

Basically, when you're trying to test some hypothesis about a proportion (e.g. the proportion of red-haired people in the US or the proportion of gay Zergs), you start by assuming the proportion to equal something (this is the "null hypothesis"). Then, using a t-test, you can figure out the probability that you would get the results you did, assuming the null hypothesis to be true. If this probability is very small, you can say that the null hypothesis is false because what happened would have been very unlikely if it were true. If it's larger, then you say that you "do not reject" the null hypothesis, because it doesn't show that your hypothesis is true; it just shows that if it is true, then your results aren't too weird.

Calc is way more interesting though, in my opinion. Also prettier! I love integral signs. In multivar, most people hated when we did triple integrals in polar coordinates, but I loved it because it's so damn pretty. Seriously, I was doing homework on a plane home for Thanksgiving, and the flight attendant complemented me on how pretty my homework was. Awesome.

Gay engineers represent?

On January 13 2012 09:36 Klondikebar wrote:


I'm not an engineer but I am WAY jealous of gay engineer sex.

+ Show Spoiler +

On January 13 2012 10:03 matjlav wrote:


Haha, love SMBC. Though that comic assumes that the gay engineers (unlike me) are not too socially awkward to get a boyfriend On that note, it looks like I'll be going on my first actual date - ever - this weekend. Wish me luck, lol.

On January 13 2012 10:06 Klondikebar wrote:


LUCK WISHED! That's actually a really exciting landmark.

On January 13 2012 10:32 matjlav wrote:


Haha, yes, it should be. Unfortunately, most of what my mind says to me about that is "WHY DID IT TAKE YOU SO DAMN LONG." Ah, well. Still pretty excited though.

On January 13 2012 10:32 matjlav wrote:


Haha, yes, it should be. Unfortunately, most of what my mind says to me about that is "WHY DID IT TAKE YOU SO DAMN LONG." Ah, well. Still pretty excited though.

On January 13 2012 10:45 Klondikebar wrote:


Because gay guys have a harder time accepting that they like guys, they have a harder time being comfortable or feeling safe in public with another guy, and then statistically...it's just harder for gay guys to even find each other to ask each other out. I would venture that most, if not all, of us started dating WAY later than our straight friends.

On January 13 2012 10:50 matjlav wrote:


Yeah, but the weird thing is that I've been totally okay with it for years. I just didn't come out until the last few months mostly because I just never felt like I had a good way to break the subject, haha. Also a general dislike of changing the status quo. and I think I was always kind of nervous that I would still never get laid after coming out, and then I'd be out of excuses for myself. I dunno. Things are at least looking up at the moment.