“The Road not taken” statistically taken: A probabilistic analysis of Frost’s poem

Abstract

“The Road Not Taken” is an exemplary poem composed by Robert Frost in 1916. The focal point of the paper is a careful statistical study of this poem intensifying several features such as:

1. Whether the number of words per line follows uniform distribution or not.

2. Studying the frequency distribution of word length and fitting a suitable probability model.

3. Do the number of vowels follow a binomial or a quasi binomial model?

4. Are the number of letters per line uniformly distributed?

5. Computing the empirical probability of each alphabet in the poem.

6. Analysis of rhythm in the poem.

7. Analysis of parts of speech used in the poem.

From the statistical analysis it is found that a Zero truncated Poisson distribution is a good fit for the frequency distribution of word length. The number of words per line may be taken as distributed uniformly in the poem which is tested by Chi Square goodness of fit test. Binomial distribution turns out to be a good fit for the number of vowels per stanza in the poem. The number of letters per line are distributed uniformly in the poem which is also tested by Chi Square goodness of fit test. The poem is following the rhythmic scheme abaab cdccd efeef ghggh. The number of words between two successive rhythmic words can be trichohotomized. The proportion of the first 13 letters (A to M) is more and significantly different as compared to that of the last 13 letters (N to Z). Finally verb is the maximum used part of speech in the poem.

Subject Classification:

• 62P99

Keywords:

• Zero truncated Poisson distribution
• Binomial distribution
• Chi Square goodness of fit test