1. Welcome to Tacoma World!

    You are currently viewing as a guest! To get full-access, you need to register for a FREE account.

    As a registered member, you’ll be able to:
    • Participate in all Tacoma discussion topics
    • Communicate privately with other Tacoma owners from around the world
    • Post your own photos in our Members Gallery
    • Access all special features of the site

Anyone Know Statistics?

Discussion in 'Off-Topic Discussion' started by Mr. Brown, Dec 14, 2011.

  1. Dec 14, 2011 at 11:12 PM
    #1
    Mr. Brown

    Mr. Brown [OP] Well-Known Member

    Joined:
    Jan 17, 2010
    Member:
    #29540
    Messages:
    2,193
    Gender:
    Male
    Orange County
    Vehicle:
    White Double Cab Off Road
    Suspensiopn: Donahoe coil overs, Donahoe rear reservoir shocks, Light Racing UCA's, Wheelers AAL's, Exterior: Powder-coated black billet grill, Hella Black Magic fog lights, color matched rear bumper, color matched side mirrors, undercover tonneau cover, Interior: Alpine W203 w/ Blackbird II, Weathertech Floor-liners, blue gauge light mod, Performance: AFE Stage II Intake, Banks exhaust, single cat y-pipe, Dynomat hood-liner, NST Performance pulleys, URD MAF calibrator
    [​IMG]

    Can any of you guys help me with this problem? It;s extra credit from one of my math classes and I haven't done statistics yet. I'm working on it now, but If one of you guys knoes this stuff, I'd appreciate the help.
     
  2. Dec 14, 2011 at 11:27 PM
    #2
    JimBeam

    JimBeam BECAUSE INTERNETS!! Moderator

    Joined:
    Apr 14, 2008
    Member:
    #5966
    Messages:
    41,825
    Gender:
    Male
    First Name:
    JB
    Vehicle:
    2014 DCSB TRD 4x4
    I'd argue that there is no correct answer based solely on the teacher not defining what a "significant" difference is
     
  3. Dec 14, 2011 at 11:41 PM
    #3
    colinb17

    colinb17 If at first you don't succeed, don't try skydiving

    Joined:
    Jan 6, 2008
    Member:
    #4127
    Messages:
    18,590
    Gender:
    Male
    First Name:
    Colin
    Lusby, MD
    Vehicle:
    '07 TRD Sport 4x4
    Build thread in sig...
    you start by totaling the columns and rows. also, sum all of the given number to figure out your sample size. the numbers in the table you are given is your "observed value". using the totals, you need to calculate your "expected value".

    you then calculate [(observed-expected)^2]/expected.

    sum those for each outcome.

    you will calculate your degrees of freedom by taking the (number of rows -1) x (number of collumns - 1). <<< don't count the totals as rows or columns

    hopefully that jogs your memory enough, because that's only about 3/4 of the way through the process, and i can't quite remember the next step. i want to say you take the degrees of freedom and go to one of tour tables (T, z, Ax, etc.)

    hope that helps a little.
     
  4. Dec 15, 2011 at 4:26 AM
    #4
    Mr. Brown

    Mr. Brown [OP] Well-Known Member

    Joined:
    Jan 17, 2010
    Member:
    #29540
    Messages:
    2,193
    Gender:
    Male
    Orange County
    Vehicle:
    White Double Cab Off Road
    Suspensiopn: Donahoe coil overs, Donahoe rear reservoir shocks, Light Racing UCA's, Wheelers AAL's, Exterior: Powder-coated black billet grill, Hella Black Magic fog lights, color matched rear bumper, color matched side mirrors, undercover tonneau cover, Interior: Alpine W203 w/ Blackbird II, Weathertech Floor-liners, blue gauge light mod, Performance: AFE Stage II Intake, Banks exhaust, single cat y-pipe, Dynomat hood-liner, NST Performance pulleys, URD MAF calibrator
    Thanks, for the help. I'll have to google it. I take statistics with this teacher next semester, so I think he just wanted us to start looking into it.
     
  5. Dec 15, 2011 at 6:23 AM
    #5
    colinb17

    colinb17 If at first you don't succeed, don't try skydiving

    Joined:
    Jan 6, 2008
    Member:
    #4127
    Messages:
    18,590
    Gender:
    Male
    First Name:
    Colin
    Lusby, MD
    Vehicle:
    '07 TRD Sport 4x4
    Build thread in sig...
    oh wow, that's a pretty in depth question to give someone who hasn't started the class yet.

    it's a very good class to take, but as you see with the posts above, without having taken the class, it's nearly impossible to know what to do.

    for a question like this, (from more or a theory standpoint) you are calculating what "signifficant" means, based on your population size, and number of variables (stores and colors in this case). in the end, you will have a number that you've come up with that defines "signifficant", and a number that represents the data you are given. you compare the two, and based onthe scenario, when one is greater or less than the other, you can conclude if there is a signifficant difference.

    it sounds fairly obscure, but once you are tought the material, you'll find that statistical analysis is actually very logical in it's methods.

    Good luck in the class!
     
  6. Dec 15, 2011 at 6:33 AM
    #6
    PB65stang

    PB65stang Well-Known Member

    Joined:
    Dec 2, 2009
    Member:
    #26900
    Messages:
    1,636
    Gender:
    Male
    First Name:
    Patrick
    Kansas City
    Vehicle:
    2006 Infiniti M35x
    Doesn't "significance" in Statistics usually deal with a certain number of standard deviations from the mean (for some reason I want to say 2 standard deviations)? I took Stats 1 and 2, but I did a brain dump of a lot of the material. I do seem to remember that "significant" is not an arbitrary term - there IS a certain meaning to it.
     
  7. Dec 15, 2011 at 6:35 AM
    #7
    rydaniels

    rydaniels Well-Known Member

    Joined:
    Mar 19, 2011
    Member:
    #53308
    Messages:
    226
    Gender:
    Male
    First Name:
    Ryan
    South Jersey
    Vehicle:
    4x4 4cyl manual
    bone stock except tool box in bed
    I take a class called "Biometry" which is a statistics course with the problems being biological in nature...think instead of xboxes like in the problem...we look at for instance race and smoking status on low birth weight. At any rate dont get too caught up in that problem if you havent taken the class yet...you will more than likely use a program called JMP (people call it jump)...If you havent heard about it...its like excel on 'roids...it gives all sorts of data and graphs etc etc...your problem would likely fall into a Chi-square like mentioned with the Observed minus expected and such...you would need an alpha level....because without it who is to say how significant it is...typically i use 0.05....other than that good luck with it
     
  8. Dec 15, 2011 at 6:37 AM
    #8
    PB65stang

    PB65stang Well-Known Member

    Joined:
    Dec 2, 2009
    Member:
    #26900
    Messages:
    1,636
    Gender:
    Male
    First Name:
    Patrick
    Kansas City
    Vehicle:
    2006 Infiniti M35x
    Yeah, same thing as a "confidence interval", I think. In Business Stats, we would have said .05 was a 95% confidence interval.
     
  9. Dec 15, 2011 at 6:41 AM
    #9
    rydaniels

    rydaniels Well-Known Member

    Joined:
    Mar 19, 2011
    Member:
    #53308
    Messages:
    226
    Gender:
    Male
    First Name:
    Ryan
    South Jersey
    Vehicle:
    4x4 4cyl manual
    bone stock except tool box in bed
    we deal with hypothesis testing....and when we get the test statistic or t-test value..what have you...we compare that to an alpha value usually 0.05 sometimes to 0.01 and then we decide whether to reject our null hypothesis or fail to reject it. If we reject the null then we go ahead and make a conclusion, including a confidence interval...like we had a problem comparing femur length to skeletal height and at the end in our confidence interval we had something like "a person who is 72 inches tall would have a femur length of x<y<x...its such a long drawn out class but thank the gods the final is in 3 hrs
     
  10. Dec 15, 2011 at 8:34 AM
    #10
    pfatyol

    pfatyol Well-Known Member

    Joined:
    Nov 13, 2009
    Member:
    #25875
    Messages:
    705
    Gender:
    Male
    First Name:
    Patrick
    Southcoast MA & lakes region NH
    Vehicle:
    2010 DC Sport
    American Roll cover, fog light mod, weather tech, marathon seat covers, twin booster seats.
    If you look at the number sold, Store is more significant than color.
    The equation for units sold (uncoded units - store number, color number) is
    Units sold = 5.0 + 4.8 * store + 1.5 * color

    With an alpha of .05 store is not statistically significant, with an alpha of .1 it is statistically significant. Color is not statistically significant at all.

    Used a DOE.

    Hope this helps.


     
  11. Dec 15, 2011 at 11:05 AM
    #11
    JaSkynyrd

    JaSkynyrd Ron F. Swanson

    Joined:
    May 26, 2011
    Member:
    #57269
    Messages:
    492
    Gender:
    Male
    Chattanooga, TN
    Vehicle:
    2001 XtraCab 4x4 V6 SR5
    I enjoy the fact that the professor tried to make it hip and relevant by making them Xbox 360 packages.

    Statistics is teh coolz!
     
  12. Dec 15, 2011 at 4:16 PM
    #12
    JCBigler

    JCBigler Well-Known Member

    Joined:
    Sep 13, 2011
    Member:
    #63473
    Messages:
    260
    Gender:
    Male
    Chouteau, OK
    Vehicle:
    Silver 2010 4WD, SR5, DC, TRD OR w/tow
    stock, Yaesu FT-8800R with a Maldol EX-107B, rear breather mod
    Ok, so wait, XBox 360s come in different color boxes? Where do I go to find one of those rare Green XBoxes? And goes it come with better games?
     
To Top