+ Reply to Thread
Results 1 to 3 of 3

Thread: Unsure Which Test to Use

  1. #1
    Points: 475, Level: 9
    Level completed: 50%, Points required for next Level: 25

    Posts
    4
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Unsure Which Test to Use




    Hi,

    We are doing some experiments in which we are poking holes of various diameters using needles (the intervals are inconsistent - 2.1 mm, 1.3 mm, 1.1 mm, and 0.9 mm) in a cylindical organ (GI tract), and then running a control fluid at room temperature through this cylinder to see at what minimum size it leaks out. We are then heating the fluid and running it through a tube to see if the minimum size has decreased. What sort of a statistical test should I use to compare the heated fluid experiments with the control at room temperature? It is not quite a continuous scale, so I'm confused.

    Thank you very much in advance!

  2. #2
    Fortran must die
    Points: 58,790, Level: 100
    Level completed: 0%, Points required for next Level: 0
    noetsi's Avatar
    Posts
    6,532
    Thanks
    692
    Thanked 915 Times in 874 Posts

    Re: Unsure Which Test to Use

    If you are testing whether the minimum size has decreased when you heat the liquid one way is logistic regression. You code the DV as decrease/does not decrease (2 levels). Then you have a dummy predictor say 1 if you heat it and 0 otherwise. The t test (the p value) will show if the difference between heated and not heated has a signficant impact on whether the DV decreases or not.

    The slope is not very intuitive. If you want to find how much heating changed the odds of being in whatever level of the DV you are maximizing (for example the decrease state if you chose to maximize that in the software) look at the odds ratio. It shows (for a dummy variable) how much the odds of being in the state you maximize increases or decreases (the average difference here between heated and not heated on these odds).

    Say you are maximising decrease. And you code the predictor dummy 1 for heated. And the odds ratio is 2. Than if you apply heat you are twice as likely on average to see a decrease than if you don't heat it.
    "Very few theories have been abandoned because they were found to be invalid on the basis of empirical evidence...." Spanos, 1995

  3. #3
    Points: 475, Level: 9
    Level completed: 50%, Points required for next Level: 25

    Posts
    4
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: Unsure Which Test to Use


    Thank you very much, noetsi! I was just intending to do t tests comparing each experiment to the control and depict my data as tables, but a regression might be a good way to integrate all the data. I hadn't thought of that.

    Thank you again for your help!

+ 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