If you happen to be a Course aficionado and a bit of a nerd, you might appreciate this “labor of love” research from friend, colleague, and fellow Course student Abelardo “Abe” Ramirez. Abe devoted some time to use statistical analysis tools and expertise (familiar from prior professional experience) and applied them to A Course In Miracles. Here are his comments, followed by the documents he generated. Enjoy! :-) The image above is a “word cloud” that Abe generated which makes the words which appear more frequently larger in the graphic.
Spoiler alert: “God” is a rather popular word in ACIM as evidenced by the histograms below! :-)
Hi Bruce,
Attached are figures that show various flavors of word frequencies in the Text. The results show whole Text frequencies as well as individual chapter frequencies.
To get these results, the script I developed:
1) tokenized the whole Text into separate words
2) removed “stopwords” that the nltk library considers uninformative, i.e., words such as I, me, our, but, and, before, after, that, who…. There are about 194 of these stop words determined by the linguists that developed nltk.
3) segregated the words into chapter bins, and counted them.The file WordCount.pdf displays straight word count figures. The first graph in this file shows the word counts for the 20 most frequent words in the whole text ( as indicated by the title above the graph “ChAll”). The next 31 graphs show the 20 most common words in each of the 31 Text chapters. The title above each graph indicates the chapter number.
The file RelativeWordCount-A.pdf also compares the word frequencies in the whole Text to the frequencies in each chapter, but does it in a different way. For example, the Ch1 God word count is divided by the whole Text God word count, and the result is multiplied by 100. So, for Ch1 graph you can see that the words God, Miracles and miracles are relatively more frequent in Ch1 than in the Text. These are another way of determining “emphasis” in each Chapter.
The file RelativeWordCount-B.pdf compares the word frequencies in the whole Text (blue dotted line) to the frequencies in each chapter (dashed red line). I had to normalize the word frequencies so that we could compare them. For example, the word count for “God” in Chapter 1 is divided by the total number of words in Ch1 and multiplied by 100. Similarly, the word count for “God” in the whole Text is divided by the total number of words in the whole Text. If you look at the graph for Ch1, you will see that the words miracle, Miracles, and miracle are relatively speaking more frequent in Ch1 than in the Text as whole. Thus, these larger frequencies suggest that Ch1 gives greater emphasis to the “miracles them” than the Text as whole. Consequently, I think of these graphs as “emphasis” plots.
Peace,
Abe
RelativeWordCount-A
RelativeWordCount-A.pdf