In order to determine changes in what has been taught at Carleton since 1867, I decided to take two approaches: by-hand and text-analysis. The by-hand approach to solving this problem entailed counting the number of course offerings per science department, as listed in the Carleton Academic Catalogs in 25 year intervals (1867, 1892, 1917, 1942, 1967, 1992, 2017). The text-analysis approach, which involved analyzing specific word frequencies within the Academic catalogs using Automator and Voyant Tools, was implemented to compare, and hopefully confirm, our “by-hand” findings.

Approach 1: “By-Hand”

To collect data on the number of science courses taught at Carleton, I downloaded Carleton’s Academic Catalogs for seven academic years, beginning with 1867-1868, and ending with 2017-2018, with catalogs spaced out in 25-year intervals. For each catalog, I looked through the course offerings and manually counted the number of courses offered per department, each year. If one course was offered Winter/Spring of a specific school year, I counted it as two courses. Additionally, I counted 6 credit courses as “one course” and 3 credit courses as “half of a course.” This method yielded the spreadsheet shown below. Finally, using the above chart, I created a line graph on Google Sheets to visualize my findings.

Number of courses offered in each science department, by academic year

Approach 2: “Text-Analysis”

After determining the raw number of science courses taught in each department at Carleton over time, I wanted to both compare my by-hand findings to a more computer-automated approach, and also account the growth of the college as a factor in the increasing number of courses offered at Carleton. To do so, I decided to analyze the relative number frequencies of specific words (corresponding to the science departments) in each of the seven Carleton Academic Catalogs analyzed in Approach 1.

I started by loading the seven Carleton Academic Catalogs into Automator, a Macintosh application that converts PDF documents into raw text files. I then uploaded these text files to Voyant Tools, a web-based reading and analysis environment, which allowed me to analyze relative frequencies of words that correlated with science departments at Carleton: “anatomy,” “botany,” “chemistry,” “geology,” “zoology,” “physics,” “biology,” and “environmental.” Under the simplified assumption that the total number of words in Carleton’s Course Catalogs directly corresponds to the growth/size of the college, by analyzing the number of frequencies of specific words, I was able visualize how and when each science department has Carleton expanded and diminished.

While Voyant offers many different visualization tools, I determined that a Stream Graph and Trend Graph were the best tools to analyze my data, because they clearly show how the relative number of frequencies for each word has changed over time.

View of Voyant Tools after uploading the seven raw text files. Trend graph is shown at center, and Stream Graph at right.

Approach 1: “By-Hand”

This first suite of graphs shows the raw number of courses in each science department in 25 year intervals at Carleton (1867-2017).

All Science Courses Offered at Carleton, 1867-2017
“Life Science” and “Environmental” Courses Offered at Carleton, 1867-2017
Non-Life Science/Environmental Science Courses Offered at Carleton, 1867-2017
Over Carleton’s history, the number of science words in the Academic Catalogs has decreased proportional to the total number of topics/subject discussed in the catalogs.

Approach 2: “Text-Analysis”

The graph below shows the number of science-related words appearing in the Academic Course Catalog for each of the years specific. For example, in 1967, science was a big topic of conversation within the Course Catalog.

The graph below shows which science departments made up the majority of the conversation within the Course Catalog for each year specified.

Finally, the line graph below shows overall trends in the relative number of frequencies of words correlating to specific science departments over time.

Changes in Scientific Verbiage

From the above graph, one can see a positive trend in the overall number of life science courses offered at Carleton since the college’s founding in 1867, however the verbiage used to describe these classes has varied dramatically over time. Notably, while anatomy and physiology classes existed in 1867, by 1892 they had been renamed to fall under the category of “biology.” Perhaps the most notable year on the graph, 1942, shows a total of zero biology classes. This unexpected drop was due to the college’s brief decision to divide biology into two fields: zoology and botany. However it is evident that by 1967, the terms “botany” and “zoology” largely fell out of popular scientific discourse, as classes previously named under these departments became more generally characterized as “biology.”

From the Space Age to Climate Change

While geology, physics, and chemistry courses have all been taught since the 19th century, and the number of courses in each department have generally followed an upward trend since 1867, the astronomy department stands out as experiencing serious decline since its peak 1942, however this was largely due to many astronomy classes being re-grouped as part of the physics department.

As depicted by the stream graph above, around WWI (1910s) there was a lag in the number of science related words in Carleton’s academic catalog, then beginning 1942 (WWII), a boom in science at Carleton began. By the space age in 1967, the academic catalog was flooded with science related words, especially physics, chemistry, and biology. After 1967, the number of science courses began to evens out again, back to about the levels in the 1930s and 1940s. Finally, likely a result of increasing concern over climate change and the future of our planet, we see the emergence of the environmental studies department at the beginning of the 21st century.

By-Hand vs. Text Analysis

Generally speaking, the graphs created by Approach 1 and 2 show similar results in terms of the growth and decline of specific science departments over time. The By-Hand approach is useful in that one can visualize how and when course offerings at Carleton began to diversify and expand. While the Text Analysis approach also shows these trends, it more-so accounts for the growth of the college, by showing the number of frequencies of specific words relative to the total number of words in the catalog. Additionally, Voyant tools creates many interactive graphs that allow the user to interact with the same data in many ways, a service that Google Sheets does not provide.

Broader than Carleton: Google NGram

Finally, to broaden the scope beyond Carleton and see if the trends in Carleton’s course offerings correspond with general word usage over time, I used Google NGram viewer, a tool which creates trend lines of word usage over time based on the parsing of thousands of Google Books .

General Takeaways:

  • The words “zoology” and “botany” follow very similar Google Ngram trend lines, and their continued decline begin around the time Carleton stopped teaching courses with their name (1950s-1960s)
  • The use of the word “biology” has been rapidly increasing since the mid-1940s, as have the number of biology classes at Carleton
  • “Physics” and “Chemistry” both reached peak usage in the 1960s, corresponding to Carleton’s boom in physics and chemistry classes
  • “Physics” is the most popular of the scientific words shown on the Google Ngram chart, however it is not the largest science department at Carleton (biology is)
  • While the use of “Environmental Studies” has been increasing since the 1980s, the phrase is not very commonly used in scientific discourse, proportional to the other scientific areas shown on the Ngram graph

If I were to continue this project I would like to do a more in-depth analysis of how the verbiage used to describe various science courses in the Academic Catalog has changed over time. Additionally, I would like to compare the number of science classes offered each year to the total number of courses offered at Carleton, rather than just presenting the data as a raw number. This approach would be more meaningful in that it would account for the growth of the college, rather than just giving a raw number of courses.