+ Reply to Thread
Results 1 to 6 of 6

Thread: Interpreting multiple regression beta values

  1. #1
    Points: 27, Level: 1
    Level completed: 54%, Points required for next Level: 23

    Posts
    3
    Thanks
    2
    Thanked 0 Times in 0 Posts

    Interpreting multiple regression beta values




    Basic question here but just want to double check my understanding..

    When I do a multiple regression I get beta values b1=0.01 and b2=0.7

    This basically means b2 has a larger effect than b1. Have I interpreted that correctly?

  2. #2
    Points: 4,358, Level: 42
    Level completed: 4%, Points required for next Level: 192

    Posts
    143
    Thanks
    3
    Thanked 37 Times in 34 Posts

    Re: Interpreting multiple regression beta values

    Yes. Unless you have a large sample size, the contribution of x1 is in fact likely to be non-significant.

  3. The Following User Says Thank You to Injektilo For This Useful Post:

    jfw13 (11-27-2014)

  4. #3
    Devorador de queso
    Points: 95,889, Level: 100
    Level completed: 0%, Points required for next Level: 0
    Awards:
    Posting AwardCommunity AwardDiscussion EnderFrequent Poster
    Dason's Avatar
    Location
    Tampa, FL
    Posts
    12,937
    Thanks
    307
    Thanked 2,630 Times in 2,246 Posts

    Re: Interpreting multiple regression beta values

    I wouldn't be so quick to jump to conclusions. Scale of the independent variables plays a huge part in whether an effect is significant or not and whether or not it has a larger impact than another variable.

    So my answer would be that it's not possible to tell and I would take Injektilo's statement that "Unless you have a large sample size, the contribution of x1 is in fact likely to be non-significant" with a grain of salt because you haven't told us anything about standard errors or sample sizes.
    I don't have emotions and sometimes that makes me very sad.

  5. #4
    Super Moderator
    Points: 13,151, Level: 74
    Level completed: 76%, Points required for next Level: 99
    Dragan's Avatar
    Location
    Illinois, US
    Posts
    2,014
    Thanks
    0
    Thanked 223 Times in 192 Posts

    Re: Interpreting multiple regression beta values

    Quote Originally Posted by jfw13 View Post
    Basic question here but just want to double check my understanding..

    When I do a multiple regression I get beta values b1=0.01 and b2=0.7

    This basically means b2 has a larger effect than b1. Have I interpreted that correctly?
    Also, when you use the terminology "beta values" are you referring standardized or unstandardized regression weights...your language and notation will be confusing to some readers because you're using the language "beta" and the notation b1 and b2 - where "beta" implies "standardized" regression weights and the notation "b1 and b2" implies "unstandardized" regression weights.

  6. #5
    TS Contributor
    Points: 17,779, Level: 84
    Level completed: 86%, Points required for next Level: 71
    Karabiner's Avatar
    Location
    FC Schalke 04, Germany
    Posts
    2,542
    Thanks
    56
    Thanked 640 Times in 602 Posts

    Re: Interpreting multiple regression beta values

    This basically means b2 has a larger effect than b1. Have I interpreted that correctly?
    In addition to what was said before, it would be interesting
    to know whether x1 and x2 are (highly) correlated.

    With kind regards

    K.

  7. #6
    TS Contributor
    Points: 22,428, Level: 93
    Level completed: 8%, Points required for next Level: 922
    spunky's Avatar
    Location
    vancouver, canada
    Posts
    2,135
    Thanks
    166
    Thanked 537 Times in 431 Posts

    Re: Interpreting multiple regression beta values


    Quote Originally Posted by Karabiner View Post
    In addition to what was said before, it would be interesting
    to know whether x1 and x2 are (highly) correlated.
    it would also be interesting to know if x1 is acting as a suppressor for x2. it is well-known that suppressor variables can have very small (close to 0), non-significant regression coefficients yet contribute the largest to the explanatory power of the model when looking at the differences between R-squared.

    i guess it's all just to say the size of a regression coefficient does not really tell you much in itself when its embedded within a larger regression model
    for all your psychometric needs! https://psychometroscar.wordpress.com/about/

+ Reply to Thread

           




Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts






Advertise on Talk Stats