Abstract: Semantic accessibility scale can be considered a fully-understood measure of natural language texts.This essay，on the basis of methods of automatic text summarization system and literary style analyses，puts forward a formula by which data-based SAS originating from English literature texts can be evaluated on line.It is by means of discussion about the corpus from The Old Man and the Sea written by Nobel winner that the conclusion testifies to the fact that no distinctiveness among different sampling ratios can be discovered.Despite the existence of domain limitation，domain simplicity and relativity of evaluation on line，SAS formula will benefit the literary critics who have access to internet-based English texts.