+ Reply to Thread
Results 1 to 3 of 3

Thread: Block design anova

  1. #1
    Points: 4,289, Level: 41
    Level completed: 70%, Points required for next Level: 61

    Posts
    4
    Thanks
    0
    Thanked 1 Time in 1 Post

    Block design anova




    I have a question about what the most appropriate anova design is for my data. I am posting here because I cannot seem to get the search function to work. I have data from an ecological restoration project. It was carried out using 6 plots, 3 restored and 3 non-restored. Restored and non-restored plots were paired. Organisms from the plots were identified and counted from 12 samples taken from each plot. I believe I can use a 2-way anova to test for differences but I am not sure if I should a two-way anova with replication, or without. If without, do I sum across the 12 samples? Is a nested anova more appropriate. Insight from anyone with experience with community ecology data is appreciated. Thanks.

  2. #2
    Points: 2,000, Level: 26
    Level completed: 40%, Points required for next Level: 150

    Location
    New Zealand
    Posts
    228
    Thanks
    3
    Thanked 48 Times in 47 Posts

    Re: Block design anova

    A two way with reps should be fine. Check that the residuals are more or less even and normal across the six plots. If the residuals are not OK, a square root transformation before the anova may make them so.

  3. #3
    TS Contributor
    Points: 1,905, Level: 26
    Level completed: 5%, Points required for next Level: 95

    Posts
    253
    Thanks
    39
    Thanked 71 Times in 62 Posts

    Re: Block design anova


    Quote Originally Posted by katxt View Post
    A two way with reps should be fine. Check that the residuals are more or less even and normal across the six plots. If the residuals are not OK, a square root transformation before the anova may make them so.
    Just going to add a little to clarify that you would transform the dependent variable before running the ANOVA if needed. You would also need to understand that the inference is no longer about the mean dependent variable across the populations, but rather, your inference is about the mean of the square root of the dependent variable across the populations.

+ Reply to Thread

           




Tags for this 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