+ Reply to Thread
Results 1 to 3 of 3

Thread: Comparing tests - Graph Help

  1. #1
    Points: 1,893, Level: 25
    Level completed: 93%, Points required for next Level: 7

    Thanked 0 Times in 0 Posts

    Comparing tests - Graph Help


    This is very basic but really struggling to find the correct terminology to find out what I want to do. I have collected data on two methods of testing the same thing e.g test 1 and test 2. Each tests has it's own inherent benefits and negatives. I want to demonstrate on a graph how often these tests agree and how often they do not. In my mind axis x test 1 positive or negative, and axis y test positive or negative. I have drawn how I invisage this looking and how it could look if I did it by hand but want to do it on the computer for a presentation. I have googled but can't find simple terminology to describe/find what I am after. Can anyone help me plot something like this on the computer or find terminology so I can find out for myself? (Please see attached image!)

    Thanks in advance
    Attached Images  

  2. #2
    TS Contributor
    Points: 14,811, Level: 78
    Level completed: 91%, Points required for next Level: 39
    Miner's Avatar
    Greater Milwaukee area
    Thanked 405 Times in 363 Posts

    Re: Comparing tests - Graph Help

    This is called a contingency table. It is typically used in a Chi-squared test.

  3. #3
    Omega Contributor
    Points: 38,334, 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: Comparing tests - Graph Help

    Given your content, I would call it a "classification table", which in principle is comparable to a contingency table.

    Are either of these tests a gold standard or do you know the true outcome status. If so, you could contrast them using a receiver operator characteristic curve. If you are just comparing the "Agreement", you may be able to look at versions of Cohen - Fleischmann agreement tests.

    Also, visually I have presented comparable data using "Mosaic Graphs" or "Bubble Graphs". Another option would be to do a scatter graph based on Ys being 0s or 1s and the X-axis the same thing - though you can add a jitter function or a little random noise to each value so that values are not all on top of each other, you could also add transparency, so you can tell if some values are overlapping a little.
    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