Sorry, I should have answered the other part of the question - yes it is possible to get such a 1 (i.e. lots of significant figures) depending on the model - I often used to have to deal with getting a significance value of 1 for my F-tests, just means it's so significant your head may explode :)