Our new Indie Games subforum is now open for business in G&T. Go and check it out, you might land a code for a free game. If you're developing an indie game and want to post about it, follow these directions. If you don't, he'll break your legs! Hahaha! Seriously though.
Our rules have been updated and given their own forum. Go and look at them! They are nice, and there may be new ones that you didn't know about! Hooray for rules! Hooray for The System! Hooray for Conforming!
ITT Casket is the stupidest stupid to ever stupid it up in D&D
Incenjucar
Registered User regular
Registered User regular
in Chatlog
Posts
There has to be a correlation between the two.
I read something, and then it has to be true facts.
I was inclined to say no but now I'm not so sure.
How big is the sample size here? If you took these statistics from millions of random people around the world, then you have a different issue.
If you had to bet money on who would get the highest score, it would be best to pick someone with a bigger foot.
If you found me that statistic, then I'd be wary, because those are completely unrelated in theory, but if you proved that across the entire human population, then you'd blow a couple minds.
Obesity is a body condition. As a body condition, it affects your body, therefore it completely makes sense for your health to be affected by that condition.
And that's why I would only use first-world stats, because that's the only place where the overwhelming number of deaths are related to your health.
Okay, the key to that: Your shoe size gets bigger as you age. On average, older kids are going to know more. Shoe size was related to AGE of tester.
Now you see the issue?
EDIT:
@Casket:
No, the point is you did not find the mechanism. You saw shoe size ~ test score. The actual answer is the amount of education effected the test score, which is tengentially related to age, which is related to shoe size.
Finding ways to increase the shoe size of children won't increase their test score.
Does it matter?
Yes.
So anyone who thought "Oh wow I guess being smart is caused by having big feet" can be shot now.
Yes, but you have not checked everyone that ever WILL live.
Nobody said that. What we said was there was a co-relation between having big feet and scoring well on tests.
Yes, and what Incenj. is saying is that it is meaningless and not useful information. It tells you nothing about the relationships of the variables in that system, because the two aren't actually related.
EDIT: i.e. one does not cause the other.
Depends really how long your willing to make your causal chain.
No, it doesn't. It doesn't mean anything, because one did not cause the other. Saying that is - and think carefully here - wrong.
Wait a minute, what the fuck am I saying
Even if they don't cause each other that doesn't make the data utterly meaningless.
Because I feel like I'm trying to explain things to a wall.
Knock that down a peg. It doesn't seem like it should be this hard to explain it to someone who's done high school.
Most of the time when people feel like they are explaining things to a wall, it's because their explanations are poor.
Hugs for Chicken dude. Bravo.
Seriously.
And that's the bottom line.
I wear a size 16 shoe and I have failed a couple of tests. That would throw the whole study of foot-to-brain study into chaos.
Then again, I have had crappy teachers and I take pre-ap classes.
There will always be exceptions to everything. We are looking at the big picture. What happens in the general cases.
Why is it wrong?
And I ask you again, what you can meaningfully say from that data? What predictions would you use it to make?
If somebody says "Here are people of various shoe sizes, and here are statistics. Bet money on which one will get the highest score." And if I had no other knowledge except for these statistics, I would follow the statistics, and bet on the bigger shoe size.
Here's a hint: nothing.
Your going to say "This data has caused us to believe there may be another cause at work here".
Okay, you're blabbering now.
In science, you start by setting up a hypothesis. You then make observations and collect data, and see if that data supports your hypothesis. If it doesn't support it to a satisfactory margin of accuracy, you change your hypothesis and start the process over again until you find a hypothesis that works.
Just because a hypothesis is proven right doesn't mean it is permanently accepted as law. Theories are proven wrong all the time. In which case, new hypotheses are made, and new data is collected.
What we are saying is that without a hypothesis, the rest doesn't come. Data is meaningless outside of the context of the scientific method.
Which would be comedic because you'd be fucked if you were looking at adult shoe sizes, or the test was written in Japanese and half the subjects were Americans the other half Japanese. In fact, the only place that data is worth anything is if you're looking at school-aged children doing a test covering material over the full range of their curriculum so they neatly get sorted approximately due to age.
In other words, you in fact know nothing. In fact, you could easily get not only no better then blind luck, you could do worse. So, in fact, that data has told you nothing.
Oh, and let's remember you said you have no other information. So, no, you don't in fact know that the data is given to you within it's only - and even then - tenuous - compliance range.
Then this isn't science. There is no hypothesis here. At least, I don't have one. We're just observing for shits and giggles. Math more than anything.
Then that is the fault of the data. Not mine.
No, there is a hypothesis.
Windbit suggested a hypothesis regarding the relationship between obesity and life expectancy.
The data he used, i.e. obesity rates in various countries and the life expectancy in those countries, is certainly not enough to support his hypothesis and draw conclusions from it.
That is what we are saying.
What you are saying is, "well, the data may not support the hypothesis, but it still means something", and we're saying no, it doesn't mean anything.
The problem is at this level you have no proof that they aren't related, but the data harvesters have gathered millions and millions of entries in tables that have shown some kind of relation between big shoe sizes and high test scores.
Basically, although you could be right, you would look like a babbling loon. Burden of proof is on you.
EDIT: And fuck it, I'm done here. Why don't you go draw some lines of best fit and then try to convince any scientist that that is useful information.
No, it's the data's fault, because I was already told the data was true before being told to make a selection.
Do not assume the experiments are fucking around and giving you false data or whatever. They are giving you the plain simple statistics that they gathered. After sitting down and looking at the statistics, and looking at the subjects, and looking at the test material (all this is given information), then you can make a good prediction.
Sure, the reasons for believing the prediction may be horribly wrong, but if the prediction comes out right anyway, then everybody wins. Jump into the crowd and surf away.
Making a prediction based on data is better than making one based on nothing at all.
For fuck's sake you stupid idiot, no you don't. You can't make a good prediction because you haven't established a mechanism for that prediction to be true. You are making a blind stab in the dark. And in fact, if you had that data, then you'd sit down and say "why the fuck would a test score be related to shoe size?" - unless apparently, the hypothetical person was you, but I'm ok with that because I'm pretty sure I can trick you into killing yourself by showing you a correlation between toddlers getting electrocuted from sticking metal into powerpoints. You're not a toddler, so you should have nothing to fear.