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.