![r pca column r pca column](https://www.researchgate.net/publication/337451732/figure/fig2/AS:828206542237696@1574471235326/a-Principal-component-analysis-PCA-score-plot-and-b-loading-column-plot-based-on-1.png)
We see that overall the females are smaller than the males. We can interpret the first component as the overall size of the turtles.
![r pca column r pca column](https://miro.medium.com/max/2000/1*ba0XpZtJrgh7UpzWcIgZ1Q.jpeg)
> s.class(pca.turtles$li,factor(Turtles))
![r pca column r pca column](http://res.cloudinary.com/dyd911kmh/image/upload/f_auto,q_auto:best/v1522841047/ggbiplot7_fh6qxn.png)
![r pca column r pca column](https://cdn.analyticsvidhya.com/wp-content/uploads/2016/03/Practical-Guide-to-Principal-Component-Analysis-PCA-in-R-Python.png)
We can easily add in the discrete categorical sex’ variable with the $call: dudi.pca(df = turtlem, scannf = F, nf = 2) > pca.turtles = dudi.pca(turtlem,scannf=F,nf=2) The following object(s) are masked from ‘package:base’: Now we compute the principal components of the data: > Turtles=read.table("/Users/susan/sphinx/data/PaintedTurtles.txt",header= T) Start with unvariate and bivariate statistics ¶