Now that we have explored how to create and curate cultural data, it’s time to start thinking about what you or other scholars might be able to do with this data. In essence, it’s time to return to a core question: why do we create data of cultural phenomena and what can we do with this data?

We have already started learning about how scholars use computational methods to analyze cultural data, and also explored the question of how much we need to understand “algorithms” to understand the arguments being made. For this group assignment, you will be starting to research and explore more in-depth how scholars have used computational methods to analyze cultural data to help you build a foundation for your use of computational methods with your group’s dataset.

Please make sure to document how you are dividing up the work for this assignment and what each group member is responsible for in detail, so that we can help you if you run into any issues and evaluate your work fairly.

In this assignment, you’ll start by trying to locate relevant scholarship that uses computational methods to study culture. This may include work from fields like Digital Humanities, Computational Social Sciences, Information Sciences, or even studies focused on collections of cultural data. Your goal is to identify and examine this scholarship with a focus on how computation is used to analyze cultural phenomena and make arguments about culture.

1. Select Methods & Data Types

For this assignment, you should explore at least three different computational methods and at least two types of cultural data. You are encouraged to consider what methods and data might be most relevant to your final project, but you are not required to use the same methods or data in your final project. Once you have collectively selected your methods and data types, you should divide up the work so that each group member is responsible for researching either a method or a data type. You are allowed to have more than one group member researching the same method or data type, but they are each responsible for locating relevant scholarship (i.e. they should not be researching the same scholarship and each person should be responsible for summarizing a different piece of scholarship).

As a group, you will need to decide both what constitutes a computational method and what constitutes a type of cultural data. For example, you might consider “topic modeling” to be a computational method and “literary texts” to be a type of cultural data. You might also consider “social network analysis” to be a computational method and “social media data” to be a type of cultural data. You should be prepared to explain why you have selected these methods and data types.

2. Search & Identify Relevant Potential Scholarship

Now that you have selected your methods and data types, you should start searching for relevant scholarship that uses these methods to analyze cultural data. The scholarship you find does not have to be extremely recent but it should be peer-reviewed and published in a reputable scholarly venue. You should specifically look for scholarship that has publicly released their data and code, as that will be part of what you review.

3. Review & Summarize Selected Scholarship

For each piece of scholarship you find, you should write a brief summary that includes the following information:

  • The bibliographic information for the piece of scholarship (author, title, publication venue, publication date, etc.).
  • A brief description of the computational method or cultural data type being used. It is alright if you do not fully understand the details of the method, but you should be able to describe the transformation of data that is happening.
  • A summary of the argument being made in the scholarship and how the computational method is used to make that argument.
  • A brief description of the code and data that is publicly available.
  • A brief assessment of what you found most interesting or useful about the scholarship and also how you would categorize it as a disciplinary approach (e.g. Digital Humanities, Computational Social Sciences, etc.).
  • Finally, you should include a brief assessment of how you think this scholarship might be useful for your group project.

4. Determine Selected Methods for Group Project

The final piece of this assignment is using this research as a foundation to decide on what methods you will be experimenting with as a group for the Experimenting with Datasets Update due in a few weeks. You should discuss as a group what you have learned from your research and collectively decide on what method makes the most sense for your dataset. You are not required to have tried this method yet, but you should be prepared to explain why you think it is the best method for your dataset, and how you determined this based on your research.

You should document your research, assessments, and selected methods in your group GitHub repository, as well as how you divided the labor and any issues you encountered. How you organize this documentation is up to you but I would highly encourage you to include links to all relevant scholarship and any screenshots or even code and data that you found useful. As a reminder, selected groups will also be presenting their findings in class on the day this assignment is due.

In addition to the resources below, you might also take a look at the list in our lesson on Exploring Computational Methods for Cultural Data.

Resources for Finding Relevant Scholarship

Selected Relevant Journals and Example Projects