I'm working with the effect of antibiotics in manure on soil bacterial communities. In my experiments, I have 2 sets of controls (bare soil and soil+manure) and 3 treatments (soil+manure+antibiotic concentration), with 4 replicates for each (total 20 microcosms). I took samples at 3 different time points (days 1, 3 and 7) to measure bacterial growth. I need to do ANOVAs for differences in growth according to treatment and sampling point, but I'm not sure how to fit my variables into parameters for analysis. Any suggestions? Thanks!
I think you can do ANOVA with repeated measure. Sampling is the repeated measure factor. If you only to measure the treatment and sampling point effect, I think you can do one-way ANOVA with repeated measure. Hope this can help you a little bit.
I agree with M. Jose Gama. You have to check the normality of data distribution (Shapiro test).
Thus, normalize the data before using an ANOVA or use a PERMANOVA. The latter has less statistical power though but does not assume a normal distribution.
You should check normality and homogeneity of variance (Levene after,e.g. Shapiro Wilk for normality). You should perform repeated measures ANOVA. Alternatively, you can integrate your measures in only one variable (e.g. growing rate) and perform ANOVA (in some cases this can be acceptable); there is a set of recommendations in Onofri et al. 2016 (Europ. J. Agronomy) for agronomy long-term experiments that can be useful for the type of experiments you performed.
Ms.Pamella,You can analyze your data using a two factor randomized complete block (RCB)design.You have five treatments(including two controls),factor a;three sampling points,factor b and four replications(r).In the ANOVA
Replications r-1 =4-1=3
Treatment ab-1=15-1=14
Antibiotic concentrations and controls(A) a-1 =5-1= 4
Sampling periods(B) b-1=3-1= 2
AxB ( a-1)(b-1) = 8
Error ( r-1)(ab-1)=42
Total rab-1 =59
Hope that this will serve your purpose and rest of analysis you can do.
- Badges
- Science topic
- Similar topics
- Psychology