+ Reply to Thread
Results 1 to 3 of 3

Thread: Visual display of one-tailed tests

  1. #1
    Points: 4, Level: 1
    Level completed: 7%, Points required for next Level: 46

    Posts
    1
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Visual display of one-tailed tests




    I am trying to show the results of an analysis with one tailed tests in a figure.

    The model uses logistic regression, with a interval variable as the independent variable, and I generate predicted probabilities. I have a directional hypothesis, so I am using a one-tailed significance test. This generally gives a higher level of statistical significance.

    The problem is that if I include 95% confidence interval around the predicted probabilities (ie the shaded area around the line), it implies different results than the one-tailed test. With a two-tailed test, using 95% CIs around the line of predicted probabilities works perfectly, but I can't figure out how to do that when using a one-tailed test.

    I could shrink the confidence intervals, but I'm not sure if that the right answer. Any suggestions?

  2. #2
    Devorador de queso
    Points: 95,819, Level: 100
    Level completed: 0%, Points required for next Level: 0
    Awards:
    Posting AwardCommunity AwardDiscussion EnderFrequent Poster
    Dason's Avatar
    Location
    Tampa, FL
    Posts
    12,935
    Thanks
    307
    Thanked 2,629 Times in 2,245 Posts

    Re: Visual display of one-tailed tests

    What do you mean by "if I include 95% confidence interval around the predicted probabilities (ie the shaded area around the line), it implies different results than the one-tailed test"?
    I don't have emotions and sometimes that makes me very sad.

  3. #3
    Omega Contributor
    Points: 38,334, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    6,998
    Thanks
    398
    Thanked 1,186 Times in 1,147 Posts

    Re: Visual display of one-tailed tests


    Can you provide more information on what you mean by interval variable; range, how many levels, etc.?

    I am guessing you are trying to create a figure that has the predicted probability or odds ratios for the "interval" variable, with one of those shaded regions around it. I have not see a version of what you are talking about. The 95% CI is still the 95% CI, so I think you may be best served by as you alluded to, using perhaps the 90% CI, but be very descriptive so the reader is not mislead. Perhaps in the figure legend include a decent description.

    What program are you using?
    Stop cowardice, ban guns!

+ 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