[9.2] Task 9: Network Assignment Using Golden Record Curation Quiz Data

Before beginning my analysis, I discovered a discrepancy in my data: while my blog post lists ten selections, my quiz response only included nine. I found that Track 4: Pygmy Girls’ Initiation Song (Zaire) was inadvertently omitted. Interestingly, Track 4 was a niche choice selected by only three other participants, all of whom were placed in Modularity Class 2 or 3. This omission leads me to wonder whether my modularity class or my connections would have shifted if I had included this tenth track. However, I will proceed with the network data as generated for the purpose of this reflection, acknowledging this missing node in my personal dataset.

Navigating Pre-existing Biases

The curation for Task 8 is inherently shaped by the original bias of the Golden Record list. Out of 27 tracks, 10 are European, while South America and Africa are represented by only two tracks each. To analyze the representation and selection of our class, I calculated the preference-to-supply ratio to see how our class’s choices deviated from the original distribution.

Selection Intensity (B/A): A value above 1.0 indicates that the track was selected more frequently than its original share. A value below 1.0 indicates the track was avoided or less preferred relative to its availability.

I observed a clear shift: European tracks were relatively under-selected compared to the original list, while Asian and North American tracks saw a significant increase in selections. This suggests that many students consciously attempted to break the Eurocentric bias of the original record to curate a more diverse global selections, exercising their agency as music curators. However, it is worth noting that within Asia, Chinese and Indian tracks received the highest votes (13 and 11, respectively). I wonder if this reflects a different kind of bias, cultural familiarity within British Columbia or a global context, rather than a purely objective curation. Conversely, the already small share of African and South American music decreased further in our selection, raising the question of whether musical unfamiliarity acted as a barrier to inclusion.

The Popular and the Marginalized

The Most Popular: Track 24 (Flowing Streams)

Track 24: Flowing Streams was chosen by 13 participants. While I expected a famous Western classical piece like the Brandenburg Concerto (which tied for 2nd) to be the most preferred, the high preference for Track 24 might be a statistical outcome of supply vs choice. With only five Asian tracks available, those seeking diversity naturally gravitated toward this piece. Interestingly, Group 1 consists entirely of those who did not choose this track. I investigated their blogs to find the reasons why they didn’t select this piece, but I couldn’t find any explicit justification.

The Least Popular: Track 10 (Gavotte en rondeaux)

Track 10: Gavotte en rondeaux received only two votes. While we can assume this Bach piece was overlooked because multiple Bach tracks were available, the data cannot explain why these two individuals felt this particular piece was essential to their narrative. Upon reviewing their blogs, I found that their stated reasons for selection shared no significant connection. Yet, despite these differing internal logics, they share three common tracks and are clustered together in Group 3.

A Deep Dive into Group 4

As a member of Group 4,  the only group with five members and the most densely connected cluster, I examined our network. While we appear as the most unified community in the visualization, our intentions were diverse:

• Sooyoung Y (Me): Focused on geographic balance, chronological evolution, and gender diversity.
• Jason L: Focused on reflecting human technological progression.
• Clarie J: Reflected harmony between nature and humanity, influenced by a background as a cellist.
• Eric W: Aimed for maximum variety in style and language, even at the cost of excluding personal favorites.
• Joy Y: (Intentions unknown, as I was not able to find her blog post).

The visualization captures surface-level commonalities but erases our individual logic. Looking only at the graph, one might assume we shared a single rationale. This highlights a political implication of data grouping: the loss of individual agency. We are connected by the what, but we are disconnected by the why. Furthermore, the placement of Joy in our group remains a mystery; Joy shares no connections with the other four members on several tracks, yet remains clustered as group 4. In fact, Joy selected Track 16 along with Megan F. and Ellie K., both of whom are in Group 1. Based on shared track selections, Joy appears more closely linked to Group 1 except Track 24, the most popular track. This leads me to question the algorithm’s grouping criteria: was the primary requirement for Group 1 simply the exclusion of the highly popular Track 24?

Reflection

This network analysis is a powerful tool for showing modularity, but it is silent regarding intentions. The links and nodes reveal patterns of what, but they hardly capture the criteria of selection. Most importantly, this data fails to interpret the reasons why certain tracks were not chosen.

The 17 tracks I did not choose carry as much of my identity and values as the 10 I did. Was a track excluded because it was redundant? Because it was too unfamiliar? The visualization treats a non-choice as a void, yet in the act of curation, exclusion is also a deliberate and political act. In the end, these visualizations show us the communities we have inadvertently formed, but they remind us that the most profound data, the human reasoning behind the click, may remain invisible to the algorithm.

Gemini (Google, 2026) was used to refine the grammar, spelling, and overall clarity of this post.