Changes in the Coefficients of Zipf’s Law for English Corpora of Different Contexts

Abstract

The statistical behavior of languages has been of great interest to linguists since the mid-20th century. The frequency distribution of words, often modeled with a probability mass function called Zipf’s Law (Zipf, 1936, 1949), is a particular target of research that has undergone increasingly intense scrutiny over the last decade. It turns out that there is an interesting gap in the research of Zipf’s Law pointed out by Steven Piantadosi – namely, language is not static and changes from context to context, and there is comparatively little examination of these fluctuations using Zipf’s Law (2014). This paper will set out a course for examining how language changes between different domains of time and different types of written and spoken media via comparing the parameters of best fit for the data using Zipf’s Law.