Assignment Title: Visualising and Narrating the Quantified Self
Name: ___Ewan O’Sullivan_____________________
Online Presentation URL: _______https://ewanosullivan.com/digital-humanities/assignment-2-visualising-and-narrating-the-quantified-self/_________________
| Criteria | Weight (%) | Self-Evaluation | Score (Out of 10) |
| 1. Organisation | 10% | – Is the structure logical with a clear introduction, body, and conclusion? | |
| – Are paragraphs cohesive and transitions smooth? | |||
| Comments: | 6% | ||
| 2. Grammar & Clarity | 10% | – Is the language clear, concise, and free from spelling or grammatical errors? | |
| – Does the writing style effectively convey the narrative? | |||
| Comments: | 6% | ||
| 3. Creativity | 20% | – Did you approach data collection and visualisation in a unique or innovative way? | |
| – Is the narrative engaging and thought-provoking? | |||
| Comments: | 12% | ||
| 4. Connection Between Visualisations & Text | 20% | – Do the visuals support and enhance the textual narrative? | |
| – Does the narrative explain the significance of the visuals effectively? | |||
| Comments: | 12% | ||
| 5. Demonstrated Use of Information Visualisation Tools | 20% | – Did you effectively use appropriate tools for data visualisation? | |
| – Have you applied visualization principles discussed in lectures? | |||
| Comments: | 12% | ||
| 6. Depth of Thought | 20% | – Have you reflected critically on your data and findings? | |
| – Did you uncover meaningful insights, patterns, or trends? | |||
| Comments: | 12% |
Additional Reflection Questions
- Data Collection
- Was your data collection process consistent and thorough?
- Did you face any challenges (e.g., missing data, technical issues), and how did you address them?
- Analysis and Visualisation
- Did you explore multiple tools and methods to analyze and visualize your data?
- Are your visuals clear, accurate, and engaging?
- Narrative Construction
- Did your narrative effectively combine textual and visual elements to tell a cohesive story?
- Were the insights and trends you uncovered meaningful and well-communicated?
- Connection to Broader Contexts
- How does this assignment deepen your understanding of the quantified self?
- Did the experience prompt you to think differently about personal data, visualization, or digital storytelling?
Self-Assigned Grade (Out of 100): ___60%________
Signature: _____Ewan O’Sullivan___________________
Date: _______11/04/2025_________________
Department of English & Digital Humanities
DH2002: Digital Tools & Methods II
Assignment #2 – Visualising and Narrating the Quantified Self
Ewan O’Sullivan
11th April, 2025
Lecturer: Dr. Shawn Day
Introduction
For my data-driven textual narrative self-analysis, I decided to track my progress learning German. This was a compelling choice for a myriad of reasons: firstly, language acquisition is a dynamic, complicated process with many challenges and, contrastingly, immense dividends. The essence of learning language is the formation of a positive habit, and an associated personal system of dedicating time and effort to the task at hand. Additionally, this undertaking has a personal edge: I will be studying in Leipzig on the Erasmus Program in 3rd Year. There are few better motivators for learning a language, than the existential dread of being stranded linguistically in the middle of another country for 10 months. My analysis will focus on day-time allocation, and track the use of the various media used to absorb the German language, identifying the triumphs and pitfalls inherent in the journey and the progress made overall.
The Data
Data was collected using Google Sheets and exported as a CSV (Comma Separated Values) file. The raw data itself spans 6 weeks (5 columns x 45 rows, or 225 distinct points of data), documenting the amount of time spent studying, the medium of language acquisition and on what date/day/week the specific session occurred.
A 0 and associated rest in the format column indicates a skipped day with no language acquisition. These were originally omitted, as I didn’t think they were important, but when I started working with the data, I realised that this was a mistake. All of the figures, trends and visualisations I produced were technically misleading, since the silences and lapses in the project spoke their own volumes about the actual learning process: categorically, no one is 100% consistent all the time at anything (least of all me), and the parts with nothing to measure still bear massive relevance to the pattern we’re trying to understand. In addition, the use of pomodoros and work timers leads to the reading times being increments of 5, where other media tend to be more dynamic and less predictable.
General Shape
A basic line plot analysis of learning time allocation over the span of all dates recorded shows a fluctuating graph with a multitude of remarkable phenomena. Graphs and visualisations are produced using the Pandas library in Python.
#1: Two steep peaks, outliers in comparison to the rest of the data
#2: Peaks and valleys occurring on a seemingly consistent basis
#3: A period of extended rest, with no activity, that precedes one of the aforementioned spikes
#1: Do(n’t) Panic?
The data is seen to skyrocket at 2 distinct points in time, in the ballparks of the 13th and the 29th of March. Seemingly, something had to have occurred on both of these days that spurred a spike in German learning. What happens if we look closer at these approximate periods of time?
Pandas allows us to create graphs within specific parameters that display certain information. In this instance, filtering by specific weeks. If we indulge a further examination of weeks 2 and 4, we can observe these jumps to pertain to the individual dates of the 13th and the 28th of March. In both instances, the numbers jump from amounts in the 10’s and 30’s of minutes up to 60 in week 2, and a cumulative total of close to 100 in Week 4. These are monumental shifts in the pattern of the data. What could be the catalyst for these changes?
Erasmus information sessions. Usually, correlation does not imply causation. But equally, the aforementioned existential dread is a terrific motivator when there’s something you simply must do to even have a chance of surviving in Germany. All this to say, it’s probably not a coincidence that I “chose” these times specifically to invest heavily in my language learning. It goes to emphasise that a significant factor is the success of any habit is purpose and intent; any meaningful effort to attain a skill must be buttressed by an objective or an ideal of success in mind. Simple reminders can stimulate vast commitments of time, especially when it relates to supporting a tectonic environmental shift in future life. It leads us to consider: at times did I do my work, and why?
#2 Thank god it’s….Monday?
Those 2 jumps are fairly anomalous in the context of the bigger picture, but the graph itself is still full of minor peaks and valleys. The lack of uniformity amongst the data points and the consistency of these shifts suggests there are smaller trends within the numbers worth discussing. Despite the indomitable human spirit and the indefatigability of my willpower, there are still reasonable limits of accountability; nobody is always productive, all the time. If we consider the total learning times accumulated across specific days of the week, is there anything enriching?
This graph is, admittedly, slightly deceptive. The previously referenced Erasmus session oddities happen to occur on Friday and Thursday, drastically boosting their performances in the final chart. In spite of this, it’s notable how Tuesday and Sunday maintain the highest “ordinary” numbers. This highlights very pertinent details about habit formation: simplicity and presence. Sunday happens to be the day I check in on assignments for the week (owing to me racking up more work hours on Saturdays & Fridays), and since I’m obliged to be at my desk anyway, I’m inclined to do the work. Similarly, Tuesdays happened to be the days that DH2002 lectures happened; with this project being directly adjacent to the lecture I was already at, the proximity encouraged effort.
Mondays are an easier puzzle to solve: work hours, on top of the busiest possible college day, leave little room for any reasonable commitment of time, within the limits of human sanity.
#3: Dunning-Krugered
In much the same manner a spirit of learning can be awakened or invigorated, it can also be crushed in the most mercilessly cruel fashion. In the latter half of the graph, there’s a significant lull pierced by only a single data point, leading to 2 chasms around a remote island of activity. What could cause this to happen? To identify the providence of my despair, we need to introduce the variable that’s remained undiscussed up to now: format.
Can you spot the problem?
The Dunning-Kruger effect (Dunning, 2011) is a psychological phenomenon that describes the tendency of the incompetent to ascribe a level of proficiency to themselves that they do not actually have. Put simply, people will dabble in a low octane application of a skill or a subject, assume they know it all (“Peak of Mount Stupid”) and be shocked to calamitous lows (“Valley of Despair”) when they realise it is not nearly as easy as they thought it was.
(2011, Wikimedia Commons)
So what went wrong, exactly? Most of my early learning activity concerned videos and reading the notes produced from those videos (which is a great way to learn the basics), with little bits of music sprinkled in between for flavour. The problem is, you can easily become very familiar with the language as you’ve written and learned it, but hearing it spoken at conversational rates and lengths is orders of magnitude less straightforward. As I soon learned: I cued but one episode of Dark, and my confidence was skewered. I was resolutely Dunning-Krugered. In the subsequent week, I could only muster listening to Kraftwerk, before I returned to seriously studying the language as I had been before.
As a final note on formats, my partiality for certain media and the dominance of the reading one is best illustrated with a pie chart. In tandem with the earlier area chart, the metamorphosis of methods over time becomes more evident; as you learn a language, you branch out from language videos and written notes to music, television and real applications. These flowers are slowly budding in the last act of the data, illustrating the initial stages of a learning transformation.
Conclusion
In learning a language and documenting my experience through data, I found the endeavour mutually beneficial: my language learning benefitted from the need for recorded, measurable effort, and my data collecting benefitted from my language learning. There were a landslide of intriguing discoveries to make in what I tracked. Even if not necessarily on the scale of an assignment in my day-to-day life, I feel I can better appreciate the importance of quantification in regards to daily tasks and habits; in tracking things about ourselves, we fundamentally get closer to understanding ourselves. We base our success on measurable progress, the perception of our own development, and that incentivises us to continue improving.
Citations
- Dunning, David. 2011. “Chapter Five – the Dunning–Kruger Effect: On Being Ignorant of One’s Own Ignorance.” Edited by James M. Olson and Mark P. Zanna. ScienceDirect. Academic Press. January 1, 2011. https://www.sciencedirect.com/science/article/abs/pii/B9780123855220000056?via%3Dihub
- “File:Dunning–Kruger Effect 01.Svg – Wikimedia Commons.” 2020. Wikimedia.org. 2020. https://commons.wikimedia.org/wiki/File:Dunning%E2%80%93Kruger_Effect_01.svg