+ Reply to Thread
Results 1 to 3 of 3

Thread: Regression question

  1. #1
    Points: 5,589, Level: 48
    Level completed: 20%, Points required for next Level: 161

    Posts
    9
    Thanks
    0
    Thanked 1 Time in 1 Post

    Regression question




    I am just beginning to learn about regression and I have a couple of questions.

    What is the difference in the standardized regression coefficient and the unstandardized regression coefficient?

    Also, what is the difference in the interpretation of the prediction equation when using the standardized regression coefficient and using the unstandardized regression coefficient?

    Thank you very much for any help!!!

  2. #2
    Admin
    Points: 20,637, Level: 90
    Level completed: 58%, Points required for next Level: 213
    quark's Avatar
    Location
    Canada
    Posts
    479
    Thanks
    25
    Thanked 196 Times in 73 Posts
    Here's how to get the standardized regression coefficients: before fitting the regression equation, all variables--response and predictors--are standardized by subtracting the mean and dividing by the standard deviation. The standardized regression coefficients, then, represent the change in response for a change of one standard deviation in a predictor.

  3. #3
    TS Contributor
    Points: 17,636, Level: 84
    Level completed: 58%, Points required for next Level: 214
    JohnM's Avatar
    Posts
    1,948
    Thanks
    0
    Thanked 6 Times in 5 Posts

    Diane,

    The difference is that the standardized regression coefficient is based on converting all of the data points into z scores (or standard scores):

    z = (x - mu) / s

    which represent the number of standard deviations an individual score is from its mean (as in the normal distribution).

    The interpretation of the equation is basically the same, whether it is standardized or unstandardized, however, the with the standardized regression, you can more easily determine the percent contribution of each coefficient (independent variable) in the equation.

    Sometimes, when you have several independent variables in a regression model, and they all have widely different measurement scales, it can make the picture a bit more clear by converting everything to standard scores.

    Here's a good link: http://www.csulb.edu/~msaintg/ppa696/696regmx.htm#STANDARDIZED

    Hope this helps a bit,
    John

+ Reply to Thread

           




Similar Threads

  1. regression question
    By chrischemma in forum Statistics
    Replies: 5
    Last Post: 07-02-2009, 02:02 PM
  2. regression question
    By muggai in forum Statistics
    Replies: 0
    Last Post: 03-24-2009, 10:01 AM
  3. regression question
    By crazzzyyyy in forum Statistics
    Replies: 1
    Last Post: 11-27-2006, 06:31 PM
  4. Regression question
    By mishery in forum Statistics
    Replies: 0
    Last Post: 06-19-2006, 10:48 AM

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