+ Reply to Thread
Results 1 to 3 of 3

Thread: Simple Logistic Regression

  1. #1
    Points: 91, Level: 1
    Level completed: 82%, Points required for next Level: 9

    Thanked 1 Time in 1 Post

    Simple Logistic Regression

    Hey everyone,

    I had a question with respect to SLR.

    In Block 0: my classification table predicts 89.3%
    In Block 1: my classification table predicts 89.3%

    Here's a screen shot of the two: http://s10.postimg.org/ykqsieqt5/block1vsblock2t_1.jpg

    I am new to logistic regression so I am having difficulty believing this should be accurate.

    Thanks for your input in advance.


    P.S. Not sure if I should have posted this in the SPSS forum or not. If it belongs there then feel free to move it obviously

  2. #2
    Omega Contributor
    Points: 38,374, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Not Ames, IA
    Thanked 1,186 Times in 1,147 Posts

    Re: Simple Logistic Regression

    What is SLR? What are the blocks supposed to represent?

    Your model is a little wonky with nobody predicted not to have the outcome?
    Stop cowardice, ban guns!

  3. #3
    TS Contributor
    Points: 5,246, Level: 46
    Level completed: 48%, Points required for next Level: 104
    maartenbuis's Avatar
    Thanked 146 Times in 123 Posts

    Re: Simple Logistic Regression

    I suspect that SLR stands for Simple Logistic Regression. However, I don't know what "simple" is suposed to mean; doest that mean that there is also Difficult Logistic Regression?

    Anyhow, your result is pretty common, and it shows how useless classification tables are. If an outcome is rare, as in your case 10.7%, you can correctly predict the outcome for 89.3% by assigning everybody to the "fail" category. This is what both models have done. You would have gotten the same result if you added no variables at all. So using this criterium you could conclude that your two models are as good / as bad as a model with no explanatory variables. It is not so bad, as it is very hard for models to improve on an empty model with this metric. This means that there is something wrong with the metric and not necessarily with your model.

  4. The Following User Says Thank You to maartenbuis For This Useful Post:

    PeterFlom (12-03-2014)

+ Reply to 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