Logo Lanfrica

berbzz13/youtube-kids-network-analysis

Domain:

digital infrastructure

Record type:

dataset
Creator:
ber
Host:
Network traffic captures comparing YouTube Kids data collection behaviour in Nigeria vs. the United States — supporting empirical research on COPPA/NDPA regulatory compliance. # YouTube Kids Network Traffic Analysis — Nigeria vs. United States ## Overview This repository contains the raw network packet capture files and extracted analysis outputs from a controlled study comparing the default data collection behaviour of the YouTube Kids application across two regulatory jurisdictions: **Nigeria** and the **United States**. The captures were collected as part of a research project examining whether the YouTube Kids app applies different data collection defaults depending on the user's geographic location — and whether observed differences correlate with the regulatory requirements of the US Children's Online Privacy Protection Act (COPPA) and Nigeria's Data Protection Act (NDPA) of 2023. The full findings are to be published in the companion articles: - A Comparative Analysis of Child Data Protection Frameworks under COPPA and the Nigeria Data Protection Act - An Empirical Comparative Network Traffic Analysis of YouTube Kids in Nigeria and the United States Examining Differential Data Collection Practices ## Repository Contents | File | Description | |------|-------------| | `capture_NG.pcapng` | 15-minute packet capture — Nigerian network environment (VPN off) | | `capture_US.pcapng` | 15-minute packet capture — US network environment (ProtonVPN, US server) | | `analysis_NG.txt` | tshark-extracted DNS queries and TLS SNI fields — Nigeria session | | `analysis_US.txt` | tshark-extracted DNS queries and TLS SNI fields — US session | | `Full Data Export | unfiltered data exported for the data capturing processs| | `Methodology| document explainging the capturing's step by step process| | `Screenshots | snapshots of the different stages of the experiment| --- ## Methodology Summary All captures were collected on a single Android Virtual Device (AVD) running Android 10 (x86_64) using `tcpdump` deployed via `adb`, ensuring a controlled and reproducible environment. Each session ran for 15 minutes of standard user interaction (browsing, …