Screen Time Data Science Project (@_@)
Looking at My Own Screen Time: September vs October 2025
I wanted to see how I actually spend time on my phone. Like many students and young adults, I doomscroll, watch, and tap just enough to make me a high-functioning zombie. The problem? I had no clue how much time was actually slipping away.
This project looks at my personal Samsung screen time over two months to figure out:
- Which apps I use the most
- When I waste the most time
- How my habits affect my productivity and wellbeing
I knew I was spending a lot of time on my phone, but I didn't really know which apps were eating my time or when I'm most distracted.
- Collected my own screen time data from an app called Stayfree (Android) that tracks your phone and app usage on a month-by-month basis and lets you download it as a clean CSV file
- Used Stayfree to block all forms of short-scrolling content for one month to see what would happen
- Cleaned and organized 80+ apps into clear categories using a Python script I wrote
- Ran descriptive, diagnostic, and predictive analyses in Power BI
- Built visualizations in Power BI
- Tested a small digital wellness intervention (blocking scrolls)
- YouTube and Instagram dominate – some days 4–7 hours on these alone!
- Weekends are worse – about 30% higher usage than weekdays
- Most of my phone time is "distraction" – 63.5% of my time is pure distraction
- Productive apps barely show up – only a tiny 8.9% of the time
I tried blocking scrolling for short-form content to see if it would help. It actually massively reduced my social media time! But instead of putting my phone down and being productive, my brain just found a loophole. I ended up watching longer videos on entertainment apps instead.
| Metric | September (Scroll Allowed) | October (Static - No Scroll) | Change |
|---|---|---|---|
| Social Media Usage | 96h | 14h | -85% ↓ |
| Entertainment Usage | 56h | 108h | +93% ↑ |
So yeah… stopping short scrolling successfully killed my social media doom scrolling, but my brain just substituted it with long-form entertainment.
doom-scrolling-analysis/
├── README.md # What you're reading now
├── data/
│ ├── raw/ # My original screen time exports
│ └── processed/ # Cleaned and ready for analysis
├── src/ # Python scripts for cleaning & processing
└── visuals/ # Power BI visualizations
| Task | Tool |
|---|---|
| Cleaning & Analysis | Python (Pandas, NumPy) |
| Visualizations | Power BI |
| Version Control | Git & GitHub |
| Data Collection | Stayfree (Android) |
This pipeline:
- Cleans the raw CSV data from my Samsung A55 using Stayfree
- Organizes apps into 15+ categories
- Calculates usage stats
- Outputs a file ready for Power BI
- 1,000+ entries covering September 1 – October 31, 2025
- Columns include: App name, category, time spent, day, weekend flag, productivity type, etc.
- I categorized apps into: Social Media, Entertainment, Productivity, Health, Education, Games
- Descriptive: How much I use each app, daily and weekly patterns
- Diagnostic: Why my usage spikes, what days and apps are the culprits
- Predictive: Guessing future patterns and high-usage days
- Prescriptive: Recommendations to improve my digital wellbeing
I built a live dashboard showing:
- Daily screen time trends
- Productivity vs distraction
- Top apps and categories (YouTube hits 9.0K minutes and Instagram hits 5.3K!)
- Weekend vs weekday patterns
- The effect of my scrolling intervention
- Distraction rules my phone – 63.5% of time spent on non-productive apps
- Behavioral substitution is real – Cutting off scrolling just made me watch long videos instead
- Weekends are my weak point – 30% higher usage than weekdays
- Late evening is my danger zone – peak distraction time
- I want to say it's so over and I am cooked, but I never stood a chance
- Limit total screen time to 3–4 hours per day
- Block apps during study hours (9 AM–5 PM)
- Make phone-free zones at home (it really helps a lot)
- Swap distraction apps for productivity alternatives
- Watch Mondays and weekends carefully
- Try the "5-minute rule" before opening apps
ABDALLA NEZAR ELGAILI ELSHIEKH BIT34503 Data Science Project Universiti Tun Hussein Onn Malaysia (UTHM)
This project was my own attempt to understand my digital habits and take control of my phone use.