Doubts about how to analyze seed germination data (percentages) in a factorial scheme.

[jj.jailton23]

Hello friends! I hope you are well.
I have a problem, or it would be a doubt.

I have a data set that is about seed germination. I have two categorical variables and one variable that is germination percentages. I cannot use a parametric analysis (ANOVA) because the data do not follow normality and homogeneity. I was wondering if you guys help me on how I can analyze this data. I will put an example of what my data looks like. These data are: two seed treatments (water and KNO3) and Accesses (they are the species), so I evaluated the germination.

TRAT	ACESSOS	GER
agua	7754_04_06	0
agua	7754_04_06	0
agua	7754_04_06	0
agua	7754_04_06	0
kno3	7754_04_06	0
kno3	7754_04_06	0
kno3	7754_04_06	0
kno3	7754_04_06	0
agua	ARI_31_05	61
agua	ARI_31_05	33
agua	ARI_31_05	50
agua	ARI_31_05	74
kno3	ARI_31_05	5
kno3	ARI_31_05	7
kno3	ARI_31_05	0
kno3	ARI_31_05	2
agua	ARI_06_06	71
agua	ARI_06_06	30
agua	ARI_06_06	58
agua	ARI_06_06	30
kno3	ARI_06_06	3
kno3	ARI_06_06	0
kno3	ARI_06_06	0
kno3	ARI_06_06	0

[phil_hummel]

I would suggest starting with some exploration of the data. Something like this:

library(ggplot2)
ggplot(data = your_data, mapping=aes(x=factor(TRAT), y=GER)) +
    geom_boxplot() +
    facet_grid(~factor(ACESSOS))

your_data needs to be a dataframe and I'm assuming that the categorical variables are strings so I cast them to factors in the ggplot function call.

[ross_gilmour]

I suggest you see https://www.statforbiology.com/2019/stat_survival_germination/. The code is on github.