1. Welcome! Please take a few seconds to create your free account to post threads, make some friends, remove a few ads while surfing and much more. ClutchFans has been bringing fans together to talk Houston Sports since 1996. Join us!

Stats question

Discussion in 'BBS Hangout' started by jcantu, Feb 27, 2006.

  1. jcantu

    jcantu Member

    Joined:
    Jul 3, 2002
    Messages:
    396
    Likes Received:
    22
    Anyone know anything about stats? I know next to nothing.

    I have two patient populations and I'm doing a regression analysis on data I've collected. Assuming the data is linear, I will have to linear regression equations to explain both data sets.

    My question is how do I compare the two regressions and prove that they
    are significantly different (or insignificant)? Is there a formula or test that I use?

    Anyone? Thanks!
     
  2. rhadamanthus

    rhadamanthus Member

    Joined:
    Nov 20, 2002
    Messages:
    14,304
    Likes Received:
    596
    Look at the R^2 values. The closer it is to 1, the better the fit.
     
  3. moestavern19

    moestavern19 Member

    Joined:
    Dec 8, 1999
    Messages:
    39,003
    Likes Received:
    3,641
    Compare :

    Population Variance (std dev squared)
    Population Standard Deviation
     
  4. pradaxpimp

    pradaxpimp Member

    Joined:
    Aug 3, 2002
    Messages:
    5,025
    Likes Received:
    71
    i'm really good with calculation points per game, rebounds per game, and assists per game.

    Tell me what you need, and i can probably solve it.
     
  5. Svpernaut

    Svpernaut Member

    Joined:
    Jan 10, 2003
    Messages:
    8,446
    Likes Received:
    1,029
    I've got you covered if they are stats on the diamond.
     
  6. firecat

    firecat Member

    Joined:
    Mar 25, 1999
    Messages:
    1,090
    Likes Received:
    25

    What variables are you looking at between the two populations?

    What software are you using to run the regression analysis?
     
  7. jcantu

    jcantu Member

    Joined:
    Jul 3, 2002
    Messages:
    396
    Likes Received:
    22
    For each population, I will be collecting blood work and ploting it vs. procedure. I.E. each group will have a different regression equation for the data. It is these two regressions that I want to compare to determine if they are different or similar


    I have not run the data, this is for a proposal to do the study.
     
  8. macalu

    macalu Member

    Joined:
    May 19, 2002
    Messages:
    16,942
    Likes Received:
    836
    can you translate your question into English?
     
  9. firecat

    firecat Member

    Joined:
    Mar 25, 1999
    Messages:
    1,090
    Likes Received:
    25
    It sounds like you only have one independent variable (procedure) and that you are measuring blood work as your dependent variable.

    I assume that you are looking for differences between the groups based on what procedure you used. How many different procedures are you looking at? If there is four or less, you could just run an ANOVA to find differences between the groups.

    If you are using a program like SPSS, it's all very easy to do if you know how to enter the data correctly.
     

Share This Page