First Research Reflection: Studying a Social Media Trend

July 03, 2026 • 0 comments

Image of a plastic crate on a tile floor; in reference to Kris Caballero's first research paper, reflecting on self-publishing it and tallying all the data by hand. Fully discussed on his blog post KrisComputes!

The academic part of me really wanted to do this—all these years. I aimed to pursue research for the sake of doing it. As a result, I did, and couldn't be any happier. I wanted to place my name on something, and did just that in the form of a research paper (self-)published on March 11, 2026; as this project originally started in February 25, 2026.

I wrote, and published, an essay titled, Good Thing I Failed High School. On there, I briefly mentioned how (former) classmates mocked me for failing my classes because I was "too busy" creating and producing videos—video production being my drive that time. Even though my parents were strict about my studies, I couldn't help getting into video: a very niche hobby that time; and thanks to mega-giants, like YouTube, suddenly majority of people are producing videos. Don't get me wrong, I did care about my studies, but that didn't come later prior to graduating with an Associate's degree in 2010. In my life's history, I will forever mark 2009 as the year that changed everything for me: namely my life path. It was the year I discovered computer programming—coding—and it became my sole reason I grew to love math. Despite making another conscious effort in giving video production a try, for another 8-10 years no less, that's when I started spending more time self-studying math during the 2020-2021 lockdowns; then eventually in 2024, I decided in going back to school to major in Mathematics. I earned my Bachelor's degree in Mathematics on November 2025, was on the Dean's List, and graduated with a 3.6 GPA. All this was done from a man who knew that math had no place in my life, until spiritual signs said otherwise. I mean, heck, if it wasn't for that, I don't know if I'd still be doing video production, despite not seeing results from video projects I've done. From here, I felt happier than ever, and wanted to do more.

I didn't want to sit around with a degree on hand, not doing anything. With the job market relying on skills rather than grades, I knew I had to craft something: utilizing some of the math I've learned, both in and outside of school, I thought this was the best time to do research.

Originally, I wanted to do research on why people post pictures/videos of themselves crying; believe it or not, that's the very first idea I wanted to research. Reason I didn't was because I didn't know exactly where to implement the math for it. While on the topic of social media, one of the trends that stood out in my mind was the milk crate challenge from 2021.

Believing in the Process

The surprise wasn't much until I found plenty of videos, still available, on YouTube. I do recall some awful videos, back in 2021, where people have fallen to their injuries; some had serious medical emergencies, which, unsurprisingly, YouTube has taken down. Otherwise, there I was watching each and every compilation video, including ones that were voluntarily planned and recorded—in its entirety. The difficult part of the process was finding clips that show a full shot of the crate stacks, plus the participant(s) themselves. One source I used, and watched, was forty-minutes long; and nearly fell asleep halfway of watching it, due to how slow the participants were in attempting it (that included extra scenes talking with spectators who didn't want to parcipate—"fillers," if you will). Other than that, I could call it a time in digital history, where people congregated over a challenge for possible clout. Then again, is the participation worth the struggle?

"Equationism"

Needing a sample size, I went with 100; which means I sat and watched exactly 100 video clips, counting ones that matched my requirements. The math for this was very simple—using basic probability. One of the math models I crafted was this:

$$P_S = \frac{S}{100}$$

Using success rate, S, is the probability proportional to my chosen sample size 100. Sticking with basic probability, I implemented this:

$$P(I|F) = \frac{P(I \cap F)}{P(F)}$$

Where I have, P(I|F), is the probability of injury given failure; the set I intersecting set failures, F; and P(F) is the rate of failures.

Being that I tallied all this data by hand, watching one video after the next, I wished I wore a white laboratory coat with protective goggles. I do, actually, have a clipboard where I practice exercises on textbooks I read, so there's that. Anyway, it's things like this that really make me giddy: I'm extracting numerical data from videos...on my own, with no stress.

All thanks to Emacs' org-mode, org-tables actually, I was able to jot down all the data in my 100 sample size. As said earlier, the math I used was very easy. Plenty of the research papers I've seen prior used a lot of advanced techniques; and given that this is/was my very first research paper, I kept it simple. Whether this choice makes academics frown, or green with envy, is fine with me; I want to make my stake in this world. Because I had no mentor to guide me, that's when I resorted to Google Gemini AI (LLM) for help.

Nowadays, you get sour reactions when you mention your use of AI (dismissed, even). That attitude permeated through YouTube, until many of them have started to implement uses of AI in their workflows—thumbnails, for example. While, yes, it's good practice to mention AI use, I did just that. Its role? Cleaning up my equation labeling, suggesting a better writing tone than something that writes like a blog post, like this one, and also, recommending other properties from the challenge that I may consider. I'll be honest: when I mention, along with credited, my use of AI (LLM) in my paper, I was really nervous. The moment people, mostly entities, downloaded my paper, I was mentally preparing any feedback, criticism(s), and/or backlash that may follow. I've received none so far. Not only was disclosing it a good thing, after mulling it over for a day or two, I felt better about it. Heck, I've been writing, and blogging, for over ten years on my website, plus some essays and poems here and there, so my confidence in writing is fully secured. For this paper, however, writing in a scientific tone was what took me long; not much the math nor the data recording, alongside.

A GitHub We Will Go

While there are plenty of programming languages I could code this in, presenting the data graphically, the language I decided to use, and upload on GitHub, is in MATLAB. There's a lot to say about the language, especially with Python being the better alternative, but I dedcided to choose MATLAB. Reason being is I had a great time practicing and learning it while pursuing my Bachelor's in Mathematics starting back in 2024 (I made this clear on discussion posts that were required in my classes). Anyway, if you want to clone/download it, and see for yourself, you can click here to check it out. Also, having read about it, I did this in honor of the late, great Cleve Moler: the creator of MATLAB, and co-founder of MathWorks, along with his business partner Jack Little.

As for GitHub itself, I created my account back in 2024; I barely used it until now. The reason is because I never took the time in learning how to use it. Since I'm barely learning recently, it's a great feeling; conflicted with feelings of shame and possible slander. Why? I grew up with mentally/emotionally stricken surroundings, where harsh criticism is commonplace, let alone shaming others by means of jealousy and envy—from those who were extremely insecure. For that, it's almost like if I'm barely learning something that ought to be known in contrast to what I'm working to do, I have this discomforting vision that some expert is looking down at me and give me sour faces for not knowing; almost to the point where I'm not allowed to partake in my own contributions due to my ignorance. No wonder why subjects like math are frowned upon: when people know something well, they grow frustrated when beginners start asking questions: all presumed to be "stupid" questions. It's sad there are people like this, and have witnessed this a lot on online communities/forums. Anyhow, having posted my repository, if some egotistical expert shames me for not knowing how to properly use GitHub, so be it. I like learning things at my own pace, and am confident enough to make a contribution for the good of the project/industry. If anyone feels threatened by people wanting to learn and contribute, that very someone hasn't learned much.

This project of mine is finally posted on GitHub and I'm happy...yet nervous. The nerves are in because of, as previously mentioned, those "know-it-all" people, nitpicking newcomers, like myself, when making their presence on a famous platform, is running through my mind, hoping any feedback I get will be fine and fair. Other than that, I feel accomplished. Not sure if I'll ever be one of those who makes so much contributions, their Contributions grid is mostly dark green and such. While I pride myself as a hard worker who doesn't complain over tasks I'm assigned, I also bring attention to my health. Constant thinking hurts the brain, and often times, ideas and solutions will come when I feel rested. In addition, having a good meal, and plenty of sleep, helps me a lot. Nevertheless, I'm happy to contribute in bringing us that much farther to success.