Recoding of variables in simple regression

[thomas25]

Hey!

In my statistical book, I came across recoding the variables in simple regression. I know how to record variables in order to have the values of 0 and 1. However, in the book there is written that before recording, there is a need to execute this command:

dataset$biology <- NA.

I cannot get what this NA means in this particular command, I know that it should refer to missing values but what is the exact meaning here?

Thanks

[andresrcs]

NA stands for Not Available and what is doing is filling the entire biology variable with "null values", I suppose that the intention behind this is to preallocate memory and make code run faster.

dataset <- data.frame(x = 1:10)
dataset$biology <- NA
dataset
#>     x biology
#> 1   1      NA
#> 2   2      NA
#> 3   3      NA
#> 4   4      NA
#> 5   5      NA
#> 6   6      NA
#> 7   7      NA
#> 8   8      NA
#> 9   9      NA
#> 10 10      NA

^{Created on 2019-03-31 by the reprex package (v0.2.1.9000)}