Vanilla Forums has been nominated for a second time in the CMS Critic "Critic's Choice" awards, and we need your vote! Read more here, and then do the thing (please).
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

IncenjucarIncenjucar QA Tester -> Game ProducerSeattle, WARegistered User regular
edited August 2007 in Chatlog
*rubs his temples*

Classic example:

Foot size/Test Grade.

5-6/80%
6-7/90%
7+/100%

Does this mean that foot size=Good grades?

Think carefully now.

Incenjucar on
freefallagentad_zps635a83ed.png
«1

Posts

  • Fuzzy Cumulonimbus CloudFuzzy Cumulonimbus Cloud Registered User regular
    edited August 2007
    Incenjucar wrote: »
    *rubs his temples*

    Classic example:

    Foot size/Test Grade.

    5-6/80%
    6-7/90%
    7+/100%

    Does this mean that foot size=Good grades?

    Think carefully now.
    But.... it says it right there.
    There has to be a correlation between the two.
    I read something, and then it has to be true facts. :P

    39kEWYh.jpg
  • CasketCasket __BANNED USERS
    edited August 2007
    Incenjucar wrote: »
    *rubs his temples*

    Classic example:

    Foot size/Test Grade.

    5-6/80%
    6-7/90%
    7+/100%

    Does this mean that foot size=Good grades?

    Think carefully now.

    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.

    casketiisigih1.png
  • FirstComradeStalinFirstComradeStalin Registered User regular
    edited August 2007
    Incenjucar wrote: »
    *rubs his temples*

    Classic example:

    Foot size/Test Grade.

    5-6/80%
    6-7/90%
    7+/100%

    Does this mean that foot size=Good grades?

    Think carefully now.

    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.

    Picture1-4.png
  • IncenjucarIncenjucar QA Tester -> Game Producer Seattle, WARegistered User regular
    edited August 2007
    *sigh*

    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?

    freefallagentad_zps635a83ed.png
  • Fuzzy Cumulonimbus CloudFuzzy Cumulonimbus Cloud Registered User regular
    edited August 2007
    Incenjucar wrote: »
    *sigh*

    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?
    But... it didn't say that, so how do I believe it? D:

    39kEWYh.jpg
  • electricitylikesmeelectricitylikesme Registered User regular
    edited August 2007
    Also if you conduct a test with everyone who has ever lived, you don't need statistics because you've individually checked everyone who has ever lived.

    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.

  • CasketCasket __BANNED USERS
    edited August 2007
    ege02 wrote: »
    Casket wrote: »
    Incenjucar wrote: »
    But.... it says it right there.
    There has to be a correlation between the two.
    I read something, and then it has to be true facts. :P

    This actually works, actually. Realistic results. But there's a trick to the interpretation.

    Sample size doesn't matter. The sample size could consist of the entire population throughout time.

    Sample size matters.

    If you take it from 3 people you have conducted a shitty sample, and your data is worthless.

    If you conduct the test with everyone who has ever lived, then you can make a good amount of predictions.

    The only thing a meaningfully large sample size ensures is higher accuracy.

    It doesn't make up for uncontrolled variables.

    Does it matter?

    casketiisigih1.png
  • IncenjucarIncenjucar QA Tester -> Game Producer Seattle, WARegistered User regular
    edited August 2007
    Casket wrote: »
    Incenjucar wrote: »
    *sigh*

    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?

    Yes, and the age was related to the score on the tests.

    Yes.

    So anyone who thought "Oh wow I guess being smart is caused by having big feet" can be shot now.

    freefallagentad_zps635a83ed.png
  • CasketCasket __BANNED USERS
    edited August 2007
    Also if you conduct a test with everyone who has ever lived, you don't need statistics because you've individually checked everyone who has ever lived.

    Yes, but you have not checked everyone that ever WILL live.

    casketiisigih1.png
  • ege02ege02 __BANNED USERS
    edited August 2007
    I just explained that yes, it does matter. Go read my chicken example.

    Medopine wrote: »
    Fuck that woman going "oh god oh no!!"

    It's nature, bitch
  • CasketCasket __BANNED USERS
    edited August 2007
    Incenjucar wrote: »
    Casket wrote: »
    Incenjucar wrote: »
    *sigh*

    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?

    Yes, and the age was related to the score on the tests.

    Yes.

    So anyone who thought "Oh wow I guess being smart is caused by having big feet" can be shot now.

    Nobody said that. What we said was there was a co-relation between having big feet and scoring well on tests.

    casketiisigih1.png
  • electricitylikesmeelectricitylikesme Registered User regular
    edited August 2007
    Casket wrote: »
    Incenjucar wrote: »
    Casket wrote: »
    Incenjucar wrote: »
    *sigh*

    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?

    Yes, and the age was related to the score on the tests.

    Yes.

    So anyone who thought "Oh wow I guess being smart is caused by having big feet" can be shot now.

    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.

  • CasketCasket __BANNED USERS
    edited August 2007
    Casket wrote: »
    Incenjucar wrote: »
    Casket wrote: »
    Incenjucar wrote: »
    *sigh*

    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?

    Yes, and the age was related to the score on the tests.

    Yes.

    So anyone who thought "Oh wow I guess being smart is caused by having big feet" can be shot now.

    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.

    casketiisigih1.png
  • electricitylikesmeelectricitylikesme Registered User regular
    edited August 2007
    Casket wrote: »
    Casket wrote: »
    Incenjucar wrote: »
    Casket wrote: »
    Incenjucar wrote: »
    *sigh*

    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?

    Yes, and the age was related to the score on the tests.

    Yes.

    So anyone who thought "Oh wow I guess being smart is caused by having big feet" can be shot now.

    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.

  • CasketCasket __BANNED USERS
    edited August 2007
    Casket wrote: »
    Casket wrote: »
    Incenjucar wrote: »
    Casket wrote: »
    Incenjucar wrote: »
    *sigh*

    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?

    Yes, and the age was related to the score on the tests.

    Yes.

    So anyone who thought "Oh wow I guess being smart is caused by having big feet" can be shot now.

    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.

    casketiisigih1.png
  • ege02ege02 __BANNED USERS
    edited August 2007
    Casket, don't take this the wrong way, but, have you received any college education? Or done any scientific research?

    Because I feel like I'm trying to explain things to a wall.

    Medopine wrote: »
    Fuck that woman going "oh god oh no!!"

    It's nature, bitch
  • electricitylikesmeelectricitylikesme Registered User regular
    edited August 2007
    ege02 wrote: »
    Casket, don't take this the wrong way, but, have you received any college education? Or done any scientific research?

    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.

  • CasketCasket __BANNED USERS
    edited August 2007
    ege02 wrote: »
    Casket, don't take this the wrong way, but, have you received any college education? Or done any scientific research?

    Because I feel like I'm trying to explain things to a wall.

    Most of the time when people feel like they are explaining things to a wall, it's because their explanations are poor.

    casketiisigih1.png
  • IncenjucarIncenjucar QA Tester -> Game Producer Seattle, WARegistered User regular
    edited August 2007
    Knock that down a peg. It doesn't seem like it should be this hard to explain it to someone who's done Jr. High school.

    Hugs for Chicken dude. Bravo.

    freefallagentad_zps635a83ed.png
  • ege02ege02 __BANNED USERS
    edited August 2007
    ege02 wrote: »
    Casket, don't take this the wrong way, but, have you received any college education? Or done any scientific research?

    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.

    Seriously.

    Medopine wrote: »
    Fuck that woman going "oh god oh no!!"

    It's nature, bitch
  • CasketCasket __BANNED USERS
    edited August 2007
    Look duders, even if foot size does not cause someone to score well on tests, that doesn't make the original data meaningless. It doesn't change the fact that the study found people with footsizes of 7+ score 100%s on tests, no matter how unrelated they are. You are saying because they are not causing each other, this data is completely useless. Which is wrong.

    And that's the bottom line.

    casketiisigih1.png
  • electricitylikesmeelectricitylikesme Registered User regular
    edited August 2007
    Casket wrote: »
    Look duders, even if foot size does not cause someone to score well on tests, that doesn't make the original data meaningless. It doesn't change the fact that the study found people with footsizes of 7+ score 100%s on tests, no matter how unrelated they are. You are saying because they are not causing each other, this data is completely useless. Which is wrong.

    And that's the bottom line.
    What would you use this data for?

  • Shiekahn_boyShiekahn_boy Registered User
    edited August 2007
    Casket wrote: »
    Look duders, even if foot size does not cause someone to score well on tests, that doesn't make the original data meaningless. It doesn't change the fact that the study found people with footsizes of 7+ score 100%s on tests, no matter how unrelated they are. You are saying because they are not causing each other, this data is completely useless. Which is wrong.

    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.

    "your a moron you know that wolves have packs wich they rely on nd they could ever here of lone wolves? you an idiot and your gay, wolves have packs and are smart with tactics" - Youtube Wolf Enthusiast.
    What the fuck are you people even arguing about? Shut up.
  • CasketCasket __BANNED USERS
    edited August 2007
    Guys, just because exceptions exist does not mean the entire fucking thing is wrong and flawed.

    There will always be exceptions to everything. We are looking at the big picture. What happens in the general cases.

    casketiisigih1.png
  • ege02ege02 __BANNED USERS
    edited August 2007
    Casket wrote: »
    Look duders, even if foot size does not cause someone to score well on tests, that doesn't make the original data meaningless. It doesn't change the fact that the study found people with footsizes of 7+ score 100%s on tests, no matter how unrelated they are. You are saying because they are not causing each other, this data is completely useless. Which is wrong.

    And that's the bottom line.

    Why is it wrong?

    Medopine wrote: »
    Fuck that woman going "oh god oh no!!"

    It's nature, bitch
  • electricitylikesmeelectricitylikesme Registered User regular
    edited August 2007
    Casket wrote: »
    Guys, just because exceptions exist does not mean the entire fucking thing is wrong and flawed.

    There will always be exceptions to everything. We are looking at the big picture. What happens in the general cases.

    And I ask you again, what you can meaningfully say from that data? What predictions would you use it to make?

  • CasketCasket __BANNED USERS
    edited August 2007
    And yes, data may always be fucked up, flawed, due to the way it's collected. But thats what happens when you make predictions. Theres always a chance you could be horribly wrong, but think of what happens when we are horribly right. It's better than sitting on your hands and waiting to get perfectly good data, which may never come.

    casketiisigih1.png
  • CasketCasket __BANNED USERS
    edited August 2007
    Casket wrote: »
    Guys, just because exceptions exist does not mean the entire fucking thing is wrong and flawed.

    There will always be exceptions to everything. We are looking at the big picture. What happens in the general cases.

    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.

    casketiisigih1.png
  • electricitylikesmeelectricitylikesme Registered User regular
    edited August 2007
    Casket wrote: »
    And yes, data may always be fucked up, flawed, due to the way it's collected. But thats what happens when you make predictions. Theres always a chance you could be horribly wrong, but think of what happens when we are horribly right. It's better than sitting on your hands and waiting to get perfectly good data, which may never come.
    So answer the fucking question. Why is the data relating shoe size to test scores of any value whatsoever if it doesn't describe a direct causal relationship? What are you going to say from it?

    Here's a hint: nothing.

  • CasketCasket __BANNED USERS
    edited August 2007
    Casket wrote: »
    And yes, data may always be fucked up, flawed, due to the way it's collected. But thats what happens when you make predictions. Theres always a chance you could be horribly wrong, but think of what happens when we are horribly right. It's better than sitting on your hands and waiting to get perfectly good data, which may never come.
    So answer the fucking question. Why is the data relating shoe size to test scores of any value whatsoever if it doesn't describe a direct causal relationship? What are you going to say from it?

    Here's a hint: nothing.

    Your going to say "This data has caused us to believe there may be another cause at work here".

    casketiisigih1.png
  • ege02ege02 __BANNED USERS
    edited August 2007
    Casket wrote: »
    And yes, data may always be fucked up, flawed, due to the way it's collected. But thats what happens when you make predictions. Theres always a chance you could be horribly wrong, but think of what happens when we are horribly right. It's better than sitting on your hands and waiting to get perfectly good data, which may never come.

    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.

    Medopine wrote: »
    Fuck that woman going "oh god oh no!!"

    It's nature, bitch
  • electricitylikesmeelectricitylikesme Registered User regular
    edited August 2007
    Casket wrote: »
    Casket wrote: »
    Guys, just because exceptions exist does not mean the entire fucking thing is wrong and flawed.

    There will always be exceptions to everything. We are looking at the big picture. What happens in the general cases.

    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.

    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.

  • CasketCasket __BANNED USERS
    edited August 2007
    ege02 wrote: »
    Casket wrote: »
    And yes, data may always be fucked up, flawed, due to the way it's collected. But thats what happens when you make predictions. Theres always a chance you could be horribly wrong, but think of what happens when we are horribly right. It's better than sitting on your hands and waiting to get perfectly good data, which may never come.

    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.

    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.

    casketiisigih1.png
  • electricitylikesmeelectricitylikesme Registered User regular
    edited August 2007
    Casket wrote: »
    Casket wrote: »
    And yes, data may always be fucked up, flawed, due to the way it's collected. But thats what happens when you make predictions. Theres always a chance you could be horribly wrong, but think of what happens when we are horribly right. It's better than sitting on your hands and waiting to get perfectly good data, which may never come.
    So answer the fucking question. Why is the data relating shoe size to test scores of any value whatsoever if it doesn't describe a direct causal relationship? What are you going to say from it?

    Here's a hint: nothing.

    Your going to say "This data has caused us to believe there may be another cause at work here".
    No, I'm not. I'm going to conclude that because there's no logical reason for those two things to be related. Which, in fact, I can conclude without collecting stupid data.

  • CasketCasket __BANNED USERS
    edited August 2007
    Casket wrote: »
    Casket wrote: »
    Guys, just because exceptions exist does not mean the entire fucking thing is wrong and flawed.

    There will always be exceptions to everything. We are looking at the big picture. What happens in the general cases.

    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.

    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 that is the fault of the data. Not mine.

    casketiisigih1.png
  • ege02ege02 __BANNED USERS
    edited August 2007
    Casket wrote: »
    ege02 wrote: »
    Casket wrote: »
    And yes, data may always be fucked up, flawed, due to the way it's collected. But thats what happens when you make predictions. Theres always a chance you could be horribly wrong, but think of what happens when we are horribly right. It's better than sitting on your hands and waiting to get perfectly good data, which may never come.

    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.

    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.

    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.

    Medopine wrote: »
    Fuck that woman going "oh god oh no!!"

    It's nature, bitch
  • CasketCasket __BANNED USERS
    edited August 2007
    Casket wrote: »
    Casket wrote: »
    And yes, data may always be fucked up, flawed, due to the way it's collected. But thats what happens when you make predictions. Theres always a chance you could be horribly wrong, but think of what happens when we are horribly right. It's better than sitting on your hands and waiting to get perfectly good data, which may never come.
    So answer the fucking question. Why is the data relating shoe size to test scores of any value whatsoever if it doesn't describe a direct causal relationship? What are you going to say from it?

    Here's a hint: nothing.

    Your going to say "This data has caused us to believe there may be another cause at work here".
    No, I'm not. I'm going to conclude that because there's no logical reason for those two things to be related. Which, in fact, I can conclude without collecting stupid data.

    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.

    casketiisigih1.png
  • electricitylikesmeelectricitylikesme Registered User regular
    edited August 2007
    Casket wrote: »
    Casket wrote: »
    Casket wrote: »
    Guys, just because exceptions exist does not mean the entire fucking thing is wrong and flawed.

    There will always be exceptions to everything. We are looking at the big picture. What happens in the general cases.

    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.

    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 that is the fault of the data. Not mine.
    No, it's your fault. You're the idiot here telling us the data is true. And I'm telling you, yes, and it's worthless as a result because it depends on a very specific set of conditions establishing a very specific causal pathway. It means nothing without context, and tells us nothing about the relationship between those two variables.

    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.

  • CasketCasket __BANNED USERS
    edited August 2007
    Casket wrote: »
    Casket wrote: »
    Casket wrote: »
    Guys, just because exceptions exist does not mean the entire fucking thing is wrong and flawed.

    There will always be exceptions to everything. We are looking at the big picture. What happens in the general cases.

    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.

    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 that is the fault of the data. Not mine.
    No, it's your fault. You're the idiot here telling us the data is true. And I'm telling you, yes, and it's worthless as a result because it depends on a very specific set of conditions establishing a very specific causal pathway. It means nothing without context, and tells us nothing about the relationship between those two variables.

    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.

    casketiisigih1.png
  • electricitylikesmeelectricitylikesme Registered User regular
    edited August 2007
    Casket wrote: »
    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.

«1
Sign In or Register to comment.