1. We have added a Gift Upgrades feature that allows you to gift an account upgrade to another member, just in time for the holiday season. You can see the gift option when going to the Account Upgrades screen, or on any user profile screen.
    Dismiss Notice

Statistics....

Discussion in 'Off-Topic' started by Che Guava, Mar 30, 2006.

?

Statistics are...

  1. the devil incarnate!

    3 vote(s)
    5.2%
  2. fibs wrapped in lies!

    6 vote(s)
    10.3%
  3. useful but unpleasant...

    27 vote(s)
    46.6%
  4. Fun for the whole family

    22 vote(s)
    37.9%
  1. Urederra

    Urederra Mostly harmless

    Joined:
    Apr 22, 2004
    Messages:
    5,310
    Location:
    Sea of tranquility
    Chemistry does not need much of it (depends on the particular issue you are studying, though). For me, it is either the reaction works or not. If I can isolate or identify the product I was looking for, it works, If I can't it doesn't work.

    Medicinal Chemistry is something more complicated, since there are biological tests and you can have some variability in the data, but at the end you focus on the media and you only look at the standard deviation when the activity of the compound under testing is not so great and you are wondering if the activity data is important enough to be taken into account.

    Which brings the question:

    Is statistics only important when the phenomena under study is too weak and it is difficult to tell whether the cause under study is the responsible of the studied effects?

    Let me give you a couple of examples taken from off-topic threads.

    * Second-hand smoking.
    VRVCAgent opened this thread with the following numbers:
    Those exposed to second-hand smoke:
    12.5 deaths per 1,000,000
    Those not exposed to second-hand smoke:
    10 deaths per 1,000,000

    So, some people (CarlosMM and others) say that there is a 25 % increment, so it is highly significant. Others, like Nanocyborgasm or ChrTh, say that the increment is not significant. I don't know what the standard deviation is in this particular example, and with those small numbers (2.5 deaths in one million), I tend to think that is not a big issue, but I am not sure. What do you guys think about that?

    <EDIT>
    (I recall reading somewhere that The Lancet does not publish anything in statistical medicine with less than a two fold increase i don't know if that is true, Does anybody know if it is true?)
    </EDIT>

    * The whole global warming thing, with fractions of degrees of change and pretty big standard deviations, plus a lot of possible causes and positive and negative feedbacks between causes and effects. I am not sure about that either.

    Plus, the 100 year climate computer simulations that some people take as proven facts. That would make me laugh if they weren't using the simulations to dictate policies.
     
  2. Gothmog

    Gothmog Dread Enforcer

    Joined:
    Sep 19, 2002
    Messages:
    3,352
    My favorite course in college was statistical mechanics, an upper division physical chemistry course where one learned to derive the ideal gas law, understand the boltzman distribution, time, enthalpy and entropy, and lots of other cool stuff based on statistics and a few simple assumptions.

    Quantum mechanics is likewise based on statistical mechanics.
     
  3. Che Guava

    Che Guava The Juicy Revolutionary

    Joined:
    Apr 19, 2005
    Messages:
    5,955
    Location:
    Hali-town,
    Yeah, the only reason that everyone else needs them is because they live with the uncertainty of the real world :mischief:

    It's true! When I get a model that works just right, I get a little tingly feeling that sometimes makes me wonder if I need women at all :lol:

    Yeah, nothing like bringing up phrases like 'analysis of variance', 'eigen values', and 'poisson distribution' in everyday conversation... :mischief:

    Quite true! Nothing boils me like statistics taken out of context on the news or in political speeches.....:mad:

    It can be, but not if performed and intepreted properly. THe proper way to use stats (at least in my opinion) is to:

    i)make specific test on data for significant trends interactions. This kind of test should be used in a controlled experiment, with a well thought out model developed before-hand (eg, agricultural, medical, psychological research)

    ii) as an exploratory tool, used to look at intereactions between uncontrolled phenomena, with an open mind and as few as possible preconceived notions
    (ecology, env'tal chemistry, etc)

    Problems occur when you try to apply the wrong methods to the situation. For example, applying exploratory analysis to a controlled experiment usually means you are grabbing for straws, trying to find something that isn't there. Applying tests used for controlled experiments in an uncontrolled study usually means that you have preconceived notions about what is happenning, and you are not open to the possibility that there are other factors or interactions to consider.

    I could blab on about this for hours, so I think I'll stop myself now and try to salvage my now nerdy repuation ;)
     
  4. Taliesin

    Taliesin Puttin' on the Ritz

    Joined:
    Jun 11, 2003
    Messages:
    4,906
    Location:
    Montréal
    Ha! Nothing so fancy.

    And I take back the bit about the plebs. I like plebs.
     
  5. Grisu

    Grisu Draghetto Moderator

    Joined:
    Dec 20, 2001
    Messages:
    10,360
    Gender:
    Male
    Location:
    Switzerland
    stat can be intersting, though they are often abused to help ones point....

    just yesterday there was an article in the local newspaper about allergics and traffic deaths. the stats claimed that 30% of the swiss population are allergics. it further claimed that about 30% of all deadly traffic accidents are caused by allergics. the article concluded that according to these stats it's proven that being an allergic drastically increases the risk of causing a deadly accidend :crazyeye:
     
  6. Sidhe

    Sidhe Chieftain

    Joined:
    Jan 30, 2006
    Messages:
    12,987
    Location:
    England
    Sorry I can't find who your quoting so Im quoting your quote based on that quote, time is precious to me. ;):)

    Eigen values I would of thought aren't strictly statistics in physics though knowing the state of a particular wave based on such variables as eigen vectors or values is useless statistically as the overall wavefunction is a superposition and as such it is hardly comparable to the macro world where the cat just has to be alive or dead and can't be both. You can't make predictions based on Quantum mechanical(QM) Eigen vectors thus they are a variable applied to an equation, kind of, viewing what exists now( a number of waveforms in a superposition: all existing at once: the only non mathematical or experimental confirmation of this is the two slit experiment) they cannot be used statistically, for to have a definitive eigenvalue you would colapse the wave function and destroy the data. Thus in a long winded way I am so glad statistics aren't a big deal in physics :) We can predict what will happen in terms of probability yes, but only because we allready know how something behaves in quantum "reality". Thus the story I once linked about solving algorithms, you can solve certain algorithms by QM principles with out ever having run the computer analysis, wierd but true. Some people think it's a joke when you say you do nothing to solve a problem but it's not nothing exactly, it's just having an equation that knows the answer or can predict it before you do the mathematical steps and logical math to solve said problem. This is not in my mind pure statistics as you don't need a series of data to extrapolate future events. Just a single mathematical equation that produces a single result, not several or hundreds. Sort of taking the boredom fest out of endless graphs data sets tables and mathematical modeling techniques :lol:

    Based on statistical mechanics yes, the same as, by no means, it's far more complex than just pure predictive probability, unfortunately according to the copenhagen interpritation, it's sheer unadulterated chaos(no chance of ever knowing exactly what it will be and no chance of drawing conclusions from one experiment to another because of this :eek: something Gothhmog hates by the way.:lol:
     
  7. Che Guava

    Che Guava The Juicy Revolutionary

    Joined:
    Apr 19, 2005
    Messages:
    5,955
    Location:
    Hali-town,
    Sidhe: It was me that brought up the eigen values! they're used a lot in multivariate stats for ecology that I use (principal components, correspondence analysis)....
     
  8. onejayhawk

    onejayhawk Afflicted with reason

    Joined:
    Jul 6, 2002
    Messages:
    13,348
    Location:
    next to George Bush's parents
    You haven't seen Fantasia have you. OK, that elephant ballet, but still...

    J
     
  9. Urederra

    Urederra Mostly harmless

    Joined:
    Apr 22, 2004
    Messages:
    5,310
    Location:
    Sea of tranquility
    So... do you recongnize that not only in politics, but also in science, statistics are sometimes misused or interpreted wrongly?


    It is OK with me. Actually, I like the subject, although I don't use it very often :D
     
  10. Che Guava

    Che Guava The Juicy Revolutionary

    Joined:
    Apr 19, 2005
    Messages:
    5,955
    Location:
    Hali-town,
    Absolutely! Fortunately for scientists, we have a whole system (peer-reviewed journals) that is designed to weed out just that sort of thing. Not so much in political arenas...

    I'll send you my course term paper if you'd like ;)

    It's called "Exploring Niche Differentiation in Arctic Cladoceran Herbivores Using Correspondence Analysis"
     
  11. Rhye

    Rhye 's and Fall creator

    Joined:
    May 23, 2001
    Messages:
    9,468
    Location:
    Japan / South America
    i love statistics
     
  12. Che Guava

    Che Guava The Juicy Revolutionary

    Joined:
    Apr 19, 2005
    Messages:
    5,955
    Location:
    Hali-town,
    Why???! ;)
     
  13. Babbler

    Babbler Chieftain

    Joined:
    Sep 18, 2002
    Messages:
    5,399
    I done bits and pieces of probability and statistics in math and biology. So I know they are not just a pack of lies, but to love them? I have a whole course to say yes or no. I presume that it would be painful.
     
  14. Hotpoint

    Hotpoint Rome Treaty Legions

    Joined:
    Nov 1, 2003
    Messages:
    1,261
    Location:
    Portsmouth, Hampshire, U.K., E.U.
    Most of my work involves statistics (production and analysis of) and I took several courses in the subject at University.

    They say a statistician is someone good with numbers but who doesn't have enough personality to be an accountant :D
     
  15. Zoso318

    Zoso318 Chieftain

    Joined:
    Oct 10, 2005
    Messages:
    46
    25% of me likes statistics when they are used correctly, 25% of me hates it when they are used out of context; 1/3 of all deadly accidents in the USA someone testes positive for marijuana(I heard it in health class). Therefore it is horrible. "Umm... but really 2/3 of the accidents are caused with people who are not on pot, so isn't using pot while driving safe according to those stats? Oh and you seemed to not mention that those tests test your blood for traces of marijuana from a maximum of two weeks prior, SOOOO, therefore there is actually a 1/14 chance that person even smoked pot on that day! Please stop lying." Hah, I said that to my health teacher a couple of days ago, and she kicked me out of the class because I supposedly disrupted it.

    The other 50% of me? That percentage of me wishes to get laid. Ahh the mind of a fifteen year old is so deep aint it? :p
     
  16. RegentMan

    RegentMan Chieftain

    Joined:
    May 7, 2003
    Messages:
    6,951
    Location:
    Washington State
    I hate statistics. Calculus is where it's at.
     
  17. kingjoshi

    kingjoshi King

    Joined:
    May 28, 2002
    Messages:
    1,745
    Location:
    Columbus, OH
    You use calculus to prove a lot in probability and statistics. You shouldn't take prob and stat until you've done calculus.
     
  18. classical_hero

    classical_hero In whom I trust

    Joined:
    Jan 30, 2003
    Messages:
    33,249
    Location:
    Perth,Western Australia
    Stats can be used for just about anything. 40% of people know that. :D
     

Share This Page