+ Reply to Thread
Results 1 to 2 of 2

Thread: vision in eyes at different times

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

    Thanked 0 Times in 0 Posts

    vision in eyes at different times

    Hi everyone,

    New to the forum and am seeking a bit of guidance as to how to compare vision (visual acuity) between eyes.
    So I am trying to compare the vision of an eye when a particular complication of a disease develops - lets call this X. This complication X of the disease tends to develop in one eye at a time (first eye), and may develop in the other at sometime in the future (second eye). Both eyes would have started off with the early stages of the disease (i.e. without complication X yet) when the first eye developed complication X).
    So I have the vision from 89 people who developed this complication X in their first eye. I then followed these 89 people for a certain length of time and collected the vision when they developed complication X in their second eye, this ended up being 38 eyes of those original 89 people.

    Now I know eyes from the same person are not "independent", but I initially thought an independent t-test would be a suitable test to compare the visions between the first eyes and the second eyes because the visions were taken at different time points. However, a reviewer has told me that they should be considered "dependent" and that I should be using something like a paired t-test to compare the visions (or a Mann-whitney U test i guess) unless i could provide justification otherwise.

    So if anyone could please advise me as to whether i should be treating these as dependent or independent would (or how else i should be tackling this situation) greatly appreciated.

    Last edited by jksgoh; 03-11-2017 at 02:50 AM.

  2. #2
    Points: 1,974, Level: 26
    Level completed: 74%, Points required for next Level: 26

    New Zealand
    Thanked 48 Times in 47 Posts

    Re: vision in eyes at different times

    If vision varies naturally between subjects, then I would consider the data to be paired and a paired test like the t test or Wilcoxon would be appropriate.

+ 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