Assignment Description

For this project, I’d like you to use data collected about you to visualize some aspect of your daily experience reading or writing. This will involve making choices about where and how to collect and transform these data. It will also involve choices in identifying and visualizing patterns in those data to present to readers.

First, you’ll first need to identify an aspect for which aggregating some data might be helpful in increasing your self-knowledge. In keeping with our course theme, the aspect you choose should be of a reading or writing activity in an environment where algorithms are also acting (see Beck 135-6).

*A brief note: “reading” and “writing” are broadly conceived here. Reading can be scrolling through social media, reading Amazon item descriptions while shopping for something, or browsing Netflix for something to watch. Likewise, writing can be commenting on social media, texting, emailing, note-taking, annotating, or — as we will talk about a bit later in the course — coding. Writing also doesn’t need to be composed of alphanumeric text either: composition in other “modes” — visual, auditory, kinesthetic — also work.

_For a social media example, maybe you want to know how often you see a TikTok from a certain trend on your For You Page, and whether that corresponds to your liking activity. Or, maybe you want to know if there are times you post more often to Instagram, or comment vs. like with certain kinds of accounts.*

So, how will you get this data? You have two possibilities:

  • get some existing data about you and your habits collected by someone/something else, via an app’s GUI, data from a wearable, a data dump, an API, etc.
  • collect data yourself that captures content you are delivered or your writing activities

In either case, you will need to decide what data you will need to capture your specific aspect of interest. Then, you’ll need to decide how you will get the data that are currently available to you into a form that’s useful to you. You also may need to collect additional data from another source and join them to get at what you’re interested in.

In my Instagram example above, I’d need to know, to start, the times I commented, liked, or followed, and maybe the user. These are all available in the data dump from the platform, but maybe I need to aggregate the instantaneous times by hour to say something about my habits. I would also need to code my data according to some appropriately-specific scheme for accounts (family? friends? content creators who work for food magazines? etc.)

In the TikTok example, I would need to be clear about how I was identifying different trends (e.g., dances, tropes, a certain type of play with a sound) or loosely-defined communities (e.g., “alt tiktok,” “beans tiktok”, “engineering tik tok”). Is this something I could identify later using TikTok’s data dump, or would I have to do it in tandem with browsing?

You will then need to decide how to represent your amassed data visually. You may use a software visualization package in your scripting language of choice to produce a plot or small set of plots; or, you may draw them by hand and digitize them with a scanner or camera you may have.

I encourage you to avoid using out-of-the-box visualization solutions, such as the graphs in Microsoft Excel, since they will prove to be overly constraining if your data is appropriately individual and nuanced. That said, this is not a course on data analysis and visualization, so I’m asking you to draw from your existing skillsets when possible and learn new ones if appropriate, as needed. Please do yourself a favor and choose something that you can reasonably accomplish in a couple of weeks — I’m definitely not asking you to go learn a new programming language here. If you’re really stuck, we can hash out some possibilities together, and please get in touch sooner rather than later!

Finally, you will submit an analysis paper (1000-1250 wds.) focusing on what you learned in this process about yourself and your data. This essay should include your visualizations, and could also include any drafts that didn’t quite work out, should they be relevant to your discussion. The essay should:

  • Reflect on how your data did (or did not) illuminate how algorithms are shaping your selected aspect of a reading/writing activity.
  • Analyze the choices you made throughout the data collection/transformation and visualization processes, in terms of the choices made available to you by the data to which you had access. It should, further, consider how those choices served (or failed to serve) the ends you hoped they would meet.
  • Describe how you might rethink these choices in a hypothetical future iteration of this project.
  • Weave in concepts and readings from the course, citing where appropriate.

You are welcome to treat this either as a typical academic paper with me and your class peers as your audience, or instead as something you might consider as a blog post for a more public-facing audience; in either case, the prose can be relatively informal to the extent that it is still precise, analytical, and persuasive.

Assignment Components

  • Short Proposal (a paragraph or so; Due July 10)
    • Please type me a paragraph describing what aspect of reading/writing experience you are interested in, what data you intend to collect to capture it, where/how you intend to get it from, and what skills you possess to get you there. Please feel free to brainstorm and be provisional here, or to ask me questions.</span>
  • Analysis Paper including Visualizations (1000-1250 words; Due July 24)

Acknowledgements

This assignment was adapted from Ryan Cordell’s “Dear (My) Data” assignment for his course on “Reading and Writing in the Digital Age” at Northeastern University, and retains lots of its spirit and language.