+ Reply to Thread
Results 1 to 6 of 6

Thread: Quick question about reporting statistical results

  1. #1
    Points: 1,482, Level: 21
    Level completed: 82%, Points required for next Level: 18

    Posts
    26
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Quick question about reporting statistical results




    Hey there, just wondering about proper reporting of results of my 2x2 ANOVA experiment:

    -I compared lots of morphometric measurements of sea urchins (height, diameter, weight, etc.).
    -the two factors in the 2x2 were spine clipping and microhabitat (levels: clipped once a month vs not clipped, and living in pits vs living on flat surface)

    If I found that animals living in flat blocks had significantly greater diameter, I would not only report the F and p, but the mean and sd. My question is: if I found no significant difference, would I still report the mean and sd?

    My overall question is, should I always report the mean and sd?

  2. #2
    TS Contributor
    Points: 12,501, Level: 73
    Level completed: 13%, Points required for next Level: 349

    Posts
    951
    Thanks
    0
    Thanked 103 Times in 100 Posts

    Re: Quick question about reporting statistical results

    Yeah, report the mean and sd. It's descriptive of the urchins, and another reader might see the mean/sd and think "those are smaller than the ones I've seen around here. Maybe your sampling had some kind of bias."

  3. #3
    Points: 1,482, Level: 21
    Level completed: 82%, Points required for next Level: 18

    Posts
    26
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: Quick question about reporting statistical results

    Thanks!

  4. #4
    Points: 1,482, Level: 21
    Level completed: 82%, Points required for next Level: 18

    Posts
    26
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: Quick question about reporting statistical results

    Oh wait.
    If there is no effect of one of the factors, I just report one mean and sd, correct?

    IE: there was no effect of spine clipping on diameter.
    I just report the total mean/sd.
    not the mean/sd for clipped and also the mean/sd for non-clipped.
    right?

  5. #5
    Points: 1,482, Level: 21
    Level completed: 82%, Points required for next Level: 18

    Posts
    26
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: Quick question about reporting statistical results

    Just want to make sure I'm doing this right.

  6. #6
    TS Contributor
    Points: 12,501, Level: 73
    Level completed: 13%, Points required for next Level: 349

    Posts
    951
    Thanks
    0
    Thanked 103 Times in 100 Posts

    Re: Quick question about reporting statistical results


    I'm no expert, but I think I would want to see both reported.
    For one thing, regarding statistical tests: they are not infallible (they come with assumptions about the data, they do not give a definitive yes/no), they may fall in/out of favor over time.
    The mean/sd is a baby statistics summary that probably is less debatable, and if the groups are really equal in mean/sd then that probably eliminates the need for relying on statistical tests (but you need to do them anyway).

+ 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