Do US members upload more videos than non-US?
Company: LinkedIn
Role: Data Scientist
Category: Analytics & Experimentation
Difficulty: easy
Interview Round: Technical Screen
You suspect video posting adoption differs between US and international members.
### Tables
`members`
- `memberid` INT (PK)
- `country` VARCHAR (e.g., `'usa'`, `'uk'`, `'india'`)
- `join_date` DATE
`video_posts`
- `post_date` DATE
- `memberid` INT (FK → `members.memberid`)
- `video_length` INT — seconds
### Business question
Do **US members** upload **more videos** than **non-US members**?
### Requirements
1. Propose:
- a **primary metric** and at least one **diagnostic metric** and **guardrail**.
2. Explain how you would make a fair comparison given confounders such as:
- different member base sizes,
- different tenure (older cohorts post more),
- different observation windows.
3. Provide an analysis approach (SQL and/or Python) including:
- cohorting / normalization strategy,
- statistical testing or uncertainty estimation,
- how you would interpret results and communicate limitations.
Assume you have all of Feb 2020 data available (or specify an alternative fixed window).
Quick Answer: This question evaluates a data scientist's competency in analytics and experimentation, specifically metric design, confounder control, cohorting, and statistical inference for comparing user behavior across segments.