(note: I removed the original table because there were some mistakes in it. I preferred to use R annotation of Sal's answer.)

Hi everybody,


I got the following results for an experiment of fermentation to produce Butanol:

````
Butanol =c(11.7462,11.7904,11.9162,11.8732,11.8583,11.8677,11.8697,
11.7289,11.9296,11.7722,11.9813,11.9873,11.8058,11.8711,11.937,
11.8628,11.7786,11.7649,11.8459,11.9139,12.0537,12.1359,11.9949,
12.0752,11.9993,12.2802,12.2227,12.1274,12.1408,11.9896,12.1362,
12.265,12.1353,12.0812,12.511,12.1871,11.7881,12.1962,12.2482,12.189)

Strain = c(1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
1,1,1,1,1,1,1,1,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2)

Enzyme = c("AH","AH","AH","AH","AH","AA","AA","AA","AA","AA",
"AL","AL","AL","AL","AL","B","B","B","B","B","C","C","C","C","C",
"AA","AA","AA","AA","AA","B","B","B","B","B","C","C","C","C","C")

myData = data.frame(Butanol, Strain, Enzyme)

myData$Strain = factor(myData$Strain)
`````

- The first column is the butanol produced (outcome) in the process;

- The second column is bacteria strain since I've conducted the experiment with two different strains;

- The third column is the enzyme to accelerate the butanol production, here I have five levels: AH (high dose of enzyme A), AA (average dose of enzyme A), AL (low dose of enzyme A), B (typical dose of enzyme B) and C (typical dose of enzyme C).

- 5 replicates per treatment

Then, I ran the following code:

`anova(aov(Butanol ~ Strain*Enzyme,myData))`

Only interaction and Strain were significant, so I'm a little bit confused how to proceed.

- Am I using the correct SS Type (in my case is 1)? Firstly, I was in doubt if my data was unbalanced because I did not perform every factorial combination, so I would need to use SS Type 3.

- What should I do next? I think I should filter treatments with strain 1 and treatments with strain 2 and then, run anova for these subsets, right? Or should I run post-hoc directly for interaction? 

Again, sorry for the newbie questions. I'm giving my best to learn it.