Hi everyone,

I would like to have your opinion about the statistical test to be used to properly analyze my data.

I have monitored soil CO2 efflux from a grassland managed with three different mowing regimes:

- Mowing twice per year

- Mowing 5 times per year

- No mowing

CO2 measurements were performed bi-weekly in a complete randomized block design.

I'm interested in determining whether mowing had an effect in terms of soil CO2 emissions within the date of measurement. In NOT interested in determining if within the same treatment I have differences among the dates of measurement.

So, my idea is to perform one-way ANOVA on CO2 in each date of measurement (and one-way ANOVA on cumulative values. However, I got the suggestion to perform TWO-WAY ANOVA and consider the "date x mowing" effect.

I think that TWO-WAY ANOVA model would not fit in my case (I'm not evaluating something like the yield of 3 wheat variety with two levels of fertilization. In that case ok, I guess two-way ANOVA is the correct model).

In other words: it is true that I'm evaluating different management (mowings frequency) and it is true that mowing in each treatment has been performed on different dates, but I do not see the need to include dates as a factor in the model.

Any kind of comment or suggestion are very welcome

Best,

Matteo