My hypothesis was that the readability of the two genre's would differ, given that they cater to different audience. This hypothesis was in large part disproven.
In considering the two readability scores, 12.22 for Experimental Music and 12.44 for Pop music, they are exceedingly similar. They are also higher than I might have expected, geared towards the reading level of 12th graders. Likewise, if my hypothesis were correct, we would expect to see pop music have a lower score since its audience is more general, whereas experimental music is geared towards a more specific, and arguably more educated, audience.
In terms of their similar and dissimilar most common words, the two genres have more words in common than they do unique to each other. Again, this disproves the hypothesis that different genres would be written in different styles.
It is likely that the two genres of music were written in similar styles because they were published on the same website, and Pitchfork as a whole has a narrow readership. Pitchfork might know that people who read Pitchfork reviews for Pop music might also be inclined to be fans of Experimental music and vice versa, just by virtue of the fact that they are on a well esteemed music blog. Thus, the overlap of readership might mean all genres should be written in similar styles for consistency sake.
This project answered the question I sought out to ask. However, there are limitations. Because Pitchfork could editorialize all its content to be the same level, perhaps drawing reviews from the same source is not sufficient to answer this question with external validity. If I were to do this project again, I might sample 20 pop album reviews from 20 different music sites, likewise 20 different experimental reviews from 20 different sites, to see if music-writers, in general, right about the genres differently.
What went well in this project is that the code executes many crucial attributes that can be applied to any text file. For instance, this code can be used on books, screen plays, essays, etc. This will be helpful for calculating different types of music reviews going forward, if this project were to be revisited.
What can be improved for next time is the sample size, perhaps a greater sample size would allow for even more reliable findings. Likewise, many of the most common words for both files are common amongst any music writing (i.e., music, songs, album). Going forward, I would add those types of words to the stop-words list so that only truly unique-to-the-genre words appeared.