For this task we have decided to divide our texts in three different ways: 1) Only Hannah More texts 2) Only not signed texts 3) A combination of both Hannah More’s texts and the not signed texts. With these three groups we ran the same exact tests on all three categories with the hopes of finding some interesting results. Here are our steps:

 

1. We decided to have four different visualizations.

2. The first step is to create a folder on the computer desktop labeled "stylo" (which is the output) and inside that folder create another folder "corpus." Inside the "corpus" folder is where all the texts that you are working with go. So, for each of our three tests we had to create a new corpus file each time. We also realized that we had to save all our data for each corpus because if we didn't save it, it disappeared when we ran new tests with a new data set.

3. Mac users had to download a file named Quartz. The first step is to set your working directory and choose a CRAN mirror, we chose USA WA 1, and then go through the files and choose your “stylo” folder.

4. The next thing that happens is you have to enter in two commands to start the program:

5. Now that the program is ready the functions box comes up its called, “Stylometry with R: enter analysis parameters”. There are five separate tabs: 1. Input & Language 2. Features 3. Statistics 4. Sampling 5. Output (within each of these tabs there are several functions that you can play with or use)

6. The first choice you have to make is which type of  language setting will work for your data set. There are three option if using English and they are: English, English (contr.), and English (ALL). Our group chose to go with the first option, "vanilla" English.  From here you also choose your input format, for our group we used plain text.