Blog Post about Data Analysis

Random sampling, reproducibility, subsetting, and counting outcomes — all through a simple coin flip simulation. The concepts here apply everywhere in data analysis.

Generate synthetic data, compute descriptive stats, filter by condition, and cap outliers using winsorization — the core data cleaning toolkit in base R.

How the creative economy is evolving in 2024 — YouTube's $70B GDP contribution, declining trust in institutions, the rise of individual creators, and what data tells us about arts education outcomes.

Main effects tell you what each variable does in isolation. Interaction effects tell you whether the combination matters. Here's how to test for both using Python's statsmodels.