Am I racist? Or just a statistician?
Post Reply
 
Thread Rating:
  • 2 Votes - 3.5 Average
  • 1
  • 2
  • 3
  • 4
  • 5
05-08-2012, 11:46 AM
RE: Am I racist? Or just a statistician?
Things seem to be getting somewhat heated for some reason on this thread ever since I dropped by the other day.

Allow me to clear up what I am trying to say and we can go from there. First off let's see what the definition of statistics is according to Webster's
sta·tis·tics noun pl but singular or pl in constr \stə-ˈtis-tiks\
1
: a branch of mathematics dealing with the collection, analysis, interpretation, and presentation of masses of numerical data
2
: a collection of quantitative data
There we go. Note definition 1, statistics requires data. Can you collect data from the future? No. One can only collect data from the past (even saying the present is erroneous because the present is but a fleeting moment in time). Now, the analysis and interpretation of this data must necessarily take into account all the possible known variables. Excluding variables would lead to incorrect analyses and interpretations (I posted an article about 'Toothfairy science' somewhere that covers why this is so. It was published in "Skeptic" and can be found online too). This necessarily means that statistics are not predictors of any event but descriptors of past events.

I think the confusion largely comes from using statistics to MODEL present and future conditions/outcomes. We use data and statistics from ice cores, fossils, sediments, etc to model future climates, but the statistics themselves do not attempt to predict the outcome, the model does. And even more crucial is that the model relies on the correct analyses and interpretations of the data through proper statistics, once again, failure to include or account for all variables may lead to incorrect results whose error is compounded in the model (note that some variables may be simplified as they may not be deemed to be major contributors, perhaps such a variable in the crime statistics might be favorite sport among the criminals. There may be a statistically significant answer, but it does not necessarily have a meaningful purpose in the interpretations).

One last thing on the subject of the statistics of race in the judicial system and the analogy of being faced with 2 alleys (one full of black people and one of white people). The point is that while there may be a larger number of individuals that are black in prison and are a larger proportion of the prison population too, this is still not predictive of what the outcome will be as you walk down the alley. The statistics might say 1 in 100 black people are or have been to jail versus 1 in 1000 for white people (I made these numbers up obviously). First off we know that roughly 50% of those that are or have been to jail are repeat offenders. Statistically anyone who is arrested has a 50% chance of going back in the US versus roughly 5% in somewhere like China. The reason isn't race, it's rehabilitation. This statistic alone does not show that it is rehabilitation related, that must come from the interpretation. It also does not predict who among first offenders will be back in jail, 50/50 is about as random as you can get. Anyway, back to the alley. When faced with the choice someone will choose which alley to go down based on their comfort level and most likely based on their own race too. But using the numbers I mentioned above (1 in 100 and 1 in 1000), statistically the individual is more likely to encounter someone who has been incarcerated while walking down the alley full of black people, but it is no guarantee. Let's say there are 100 people in each alley, the odds for the black alley is 1% while the white alley is 0.1%, but (assuming this is a perfectly random situation) that guarantees nothing about what will actually happen. The individual may encounter 0 down each alley, or multiple. Ultimately it will be a random walk (http://en.wikipedia.org/wiki/Random_walk) literally.

“Science is simply common sense at its best, that is, rigidly accurate in observation, and merciless to fallacy in logic.”
—Thomas Henry Huxley
Find all posts by this user
Like Post Quote this message in a reply
[+] 3 users Like TheBeardedDude's post
05-08-2012, 11:52 AM
Am I racist? Or just a statistician?
They've done studies, you know.

http://www.youtube.com/watch?v=pjvQFtlNQ...ata_player

It was just a fucking apple man, we're sorry okay? Please stop the madness Laugh out load
~Izel
Find all posts by this user
Like Post Quote this message in a reply
[+] 1 user Likes Erxomai's post
05-08-2012, 11:55 AM
RE: Am I racist? Or just a statistician?
Basically, regardless of statistics, walking down alleyways full of lowlifes of whatever skin colour = evolutionary bad idea. Unless you've already reproduced.
Find all posts by this user
Like Post Quote this message in a reply
[+] 1 user Likes morondog's post
05-08-2012, 12:05 PM
RE: Am I racist? Or just a statistician?
(05-08-2012 11:35 AM)Erxomai Wrote:  
(05-08-2012 11:21 AM)Chas Wrote:  WTF? Are you guys denying the fact of statistics?

Or you just don't like using statistics in regard to people?

Of course I can predict that the probability distribution in a random sample of a population will likely match that of the larger population.

Oh, my mistake. I didn't realize you're a bazillionaire from your statistical predictions of the stock market.

My point is stats are just a tool. They can be right or they can be wrong. Sometimes they just describe one facet of a situation. To say statistics show more black convicts than white convicts does not make it a fact that black people are more likely to be criminals than white people. All it does is show more blacks get convicted.

But for some reason you are so enamored by your stats that you're forgetting that. Tongue

Again, WTF?
I was responding to the imprecise statements that appear to deny any predictive power of statistical data. If you deny this, quantum mechanics goes right out the window.

Skepticism is not a position; it is an approach to claims.
Science is not a subject, but a method.
[Image: flagstiny%206.gif]
Visit this user's website Find all posts by this user
Like Post Quote this message in a reply
05-08-2012, 12:19 PM
Am I racist? Or just a statistician?
(05-08-2012 12:05 PM)Chas Wrote:  
(05-08-2012 11:35 AM)Erxomai Wrote:  Oh, my mistake. I didn't realize you're a bazillionaire from your statistical predictions of the stock market.

My point is stats are just a tool. They can be right or they can be wrong. Sometimes they just describe one facet of a situation. To say statistics show more black convicts than white convicts does not make it a fact that black people are more likely to be criminals than white people. All it does is show more blacks get convicted.

But for some reason you are so enamored by your stats that you're forgetting that. Tongue

Again, WTF?
I was responding to the imprecise statements that appear to deny any predictive power of statistical data. If you deny this, quantum mechanics goes right out the window.

And I said what I said. Sorry you didn't like it. Or possibly understand it.

It was just a fucking apple man, we're sorry okay? Please stop the madness Laugh out load
~Izel
Find all posts by this user
Like Post Quote this message in a reply
05-08-2012, 12:20 PM
RE: Am I racist? Or just a statistician?
Even our understandings of statistics and quantum mechanics relies on our understanding of probabilities through statistics but that does not make them predictive, at least not in any way I would use the term. Even our understanding of electron location is based on the statistical probability of its location at any given instance in time. Statistically it is more likely for it be at the top of the bell curve, but could still be anywhere within the shells of the atom. We can't actually predict where it will be at any given time, but we would deduce that the likelihood of it being in one spot vs. another is some probability. That still does not mean that this probability is necessarily true.

And quantum mechanics does not rely on statistics being predictive, it relies on models being predictive.

“Science is simply common sense at its best, that is, rigidly accurate in observation, and merciless to fallacy in logic.”
—Thomas Henry Huxley
Find all posts by this user
Like Post Quote this message in a reply
[+] 1 user Likes TheBeardedDude's post
05-08-2012, 12:26 PM
RE: Am I racist? Or just a statistician?
(05-08-2012 12:20 PM)TheBeardedDude Wrote:  Even our understandings of statistics and quantum mechanics relies on our understanding of probabilities through statistics but that does not make them predictive, at least not in any way I would use the term. Even our understanding of electron location is based on the statistical probability of its location at any given instance in time. Statistically it is more likely for it be at the top of the bell curve, but could still be anywhere within the shells of the atom. We can't actually predict where it will be at any given time, but we would deduce that the likelihood of it being in one spot vs. another is some probability. That still does not mean that this probability is necessarily true.

And quantum mechanics does not rely on statistics being predictive, it relies on models being predictive.

OK, I see where the disconnect is.
You all are saying it is not predictive for the individual person/particle/marble/whatever. True.
I am saying it is predictive of the probability distribution of a sample. Also true.

Skepticism is not a position; it is an approach to claims.
Science is not a subject, but a method.
[Image: flagstiny%206.gif]
Visit this user's website Find all posts by this user
Like Post Quote this message in a reply
[+] 1 user Likes Chas's post
05-08-2012, 12:42 PM
RE: Am I racist? Or just a statistician?
Statistics can be used to model the distribution of a population, yes. This is why you want your sample size to be fairly large so that you are taking a more adequate census of the population you are attempting to describe.

“Science is simply common sense at its best, that is, rigidly accurate in observation, and merciless to fallacy in logic.”
—Thomas Henry Huxley
Find all posts by this user
Like Post Quote this message in a reply
05-08-2012, 02:33 PM
RE: Am I racist? Or just a statistician?
(05-08-2012 12:42 PM)TheBeardedDude Wrote:  Statistics can be used to model the distribution of a population, yes. This is why you want your sample size to be fairly large so that you are taking a more adequate census of the population you are attempting to describe.

The larger the sample, the more likely its distribution will be close to the overall one.

However, the average distribution of many small samples will also be close to the overall one. More variance sample to sample, long-term agreement.

Skepticism is not a position; it is an approach to claims.
Science is not a subject, but a method.
[Image: flagstiny%206.gif]
Visit this user's website Find all posts by this user
Like Post Quote this message in a reply
05-08-2012, 06:50 PM (This post was last modified: 05-08-2012 07:29 PM by GirlyMan.)
RE: Am I racist? Or just a statistician?
(05-08-2012 11:04 AM)Chas Wrote:  
(04-08-2012 09:38 PM)GirlyMan Wrote:  BeardedBrother's point is that those 2 words don't go together.

And BeardedBrother would be wrong. The statistics provide predictive power.

Only if you believe the problem of induction has been settled. Not everyone does.

(05-08-2012 01:52 AM)earmuffs Wrote:  
(04-08-2012 10:20 PM)GirlyMan Wrote:  Your appreciation of the word "fact" is naive and juvenile.

Your understanding for the word 'fact' is lacking.

That my friend is one fucking brilliant retort!

At least I realize it you little twerp (hugs Hug ), but you're still delusional cause you think there actually are these magical fucking factual things. (Note to self: Next time I sumo charge the little fucker with muffins for ears keep my knees closer together so he can't do a slide through kicking me in the balls.)

(05-08-2012 11:46 AM)TheBeardedDude Wrote:  I think the confusion largely comes from using statistics to MODEL present and future conditions/outcomes. We use data and statistics from ice cores, fossils, sediments, etc to model future climates, but the statistics themselves do not attempt to predict the outcome, the model does.

The model is the driver proposing a plausible mechanism of action. It's more like how well does the model accommodate the data than it is the data driving the model. When the driver can't accommodate the passengers, we just change drivers. ... BTW, that's one of the best posts I've ever read old, fat, bearded dude.

As it was in the beginning is now and ever shall be, world without end. Amen.
And I will show you something different from either
Your shadow at morning striding behind you
Or your shadow at evening rising to meet you;
I will show you fear in a handful of dust.
Find all posts by this user
Like Post Quote this message in a reply
Post Reply
Forum Jump: