Solved – Wrong degrees of freedom ANOVA in ‘R’

anovar

I am trying to analyse variance of difference between pH values measures by two different methods. I have repeated measurements from a number of subjects and I want to analyse within subject variance and between subject variance, but when I use the aov() function in R degrees of freedom seems to be incorrect.

My dataset looks like this: 

 id <- c(1,1,1,2,2,3,3,3,3,6,6,6,8,8,9,9,9,9)
 ph1 <- c(7,6,6,8,8,8,4,3,5,6,6,7,8,8,5,4,3,6) # This is the gold standard method. 
 ph2 <- c(8,6,7,8,7,8,5,4,5,6,6,7,8,9,4,3,4,6)
 df <- as.data.frame(cbind(id,ph1,ph2))

And when I calculate one-way ANOVA I get this: 

 res.aov <- aov((ph1-ph2) ~ id, data = df)
 summary(res.aov)
             Df Sum Sq Mean Sq F value Pr(>F)
 id           1  0.579  0.5792    1.17  0.295
 Residuals   16  7.921  0.4950

Should Df for between groups analysis be 5 or have I misunderstod something?

Best Answer

You forgot to factorize your id variable, ANOVA doesn't do that automatically

res.aov <- aov((ph1-ph2) ~ factor(id), data = df)
summary(res.aov)
            Df Sum Sq Mean Sq F value Pr(>F)
factor(id)   5  3.083  0.6167   1.366  0.303
Residuals   12  5.417  0.4514
Related Question