Summer Research 2025

Problem

A summer research program hosted by CIC|PCUBED at Cal State Fullerton, looking at the relationship between digital screen time and mental health. The research question was straightforward: does more time on screens track with worse mental health scores, and if so, how cleanly does that correlation actually show up in the data?

Approach

I worked in a team to clean the dataset and run a series of data science techniques on top of it — exploratory analysis, correlation work, and clustering to see whether usage patterns grouped into recognizable cohorts.

The workflow itself was structured around the program: PCUBED ran workshops on technique and tooling, and we checked in regularly with peer mentors as we hit decisions on how to clean, group, or visualize the data. Most of the time was spent on the unglamorous part — getting the data into a state where the later analysis would actually mean something.

Tech

• Python — primary language for the whole pipeline
• Jupyter — notebooks for iterative analysis and write-up
• Pandas — data cleaning and reshaping
• NumPy — numerical work
• matplotlib + seaborn — visualization, both during analysis and for the final presentation

Outcome

The headline finding lined up with intuition: more screen time correlated with worse mental health scores in the dataset. We presented our graphs, results, and process to a cohort of 40+ students at the end of the program.

The technical takeaways were real — practical Pandas, clustering, and visualization work — but the bigger ones were on the team side. This was the project where Git and GitHub stopped being abstract and started being how we actually worked: branching, reviewing, and merging without overwriting each other. Tracking hours against time sheets and scheduling work around the cohort’s pace was its own skill, separate from any of the analysis itself.