Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

Say that I have a large collection of words that I am comparing two features of them. For simplicity, I will give an example of plural and singular forms of the word.

My task is to calculate the percentage of the words whose plural form appear more than singular form in a corpus.

Everything is OK about calculating the percentage. But, obviously there would be some plural words would appear more than some others (e.g. "pages": 500000, "page": 1000, the percentage being pluralized is then 99%). On the other hand, there would be some words whose percentage might be similar, but the number of times would be much less than others (e.g. "questions": 500, "question": 1, percentage is still 99%).

What I would like to ask is that is there any statistical function which can reflect the difference between these two cases of percentage being pluralized?

share|improve this question
    
If you want to record that they are the same percentage, but different in size, you will need two descriptors - one number can't be both the same and different. So the percentage of plurals and the number of plurals would be one way. Or the percentage of plurals for "page" and the percentage of the whole text represented by the "page/pages" case would be another. –  Mark Bennet Feb 28 at 11:44

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.