Data Analysis & Visualization

R-based data analysis and visualization projects, ranging from text analysis and sports analytics to public opinion research. Most have open-source code on GitHub and an accompanying blog post.

English

Sentiment analysis of music lyrics

Part of a personal blog post which uses several musical data sources to perfom analyses of musical data on the dutch symphonic metal band, Epica. I perform tidy text analysis with different lexicons after tokenizing all lyrics from Epica’s discography, sourced from the Genius API. It involved complex data cleaning through table reshaping, record linkage and fuzzy joins.

Football statistics from Latin American regional tournaments

Part of a blog post which analyzes scraped data from FBRef and Transfermarkt from LDU Quito’s performance in the Copa Sudamericana and Libertadores, the most important Latin American football tournaments. The post includes a detailed analysis of the team’s performance in the tournaments, using data from the R worldfootballR package.

Analyzing public opinion from survey data

After a series of protests developed in Ecuador in July 2022, I analyzed public opinion variables to try to understand the public sentiment behind the protest. I implement survey design adjustments from the AmericasBarometer survey to correct for oversampling of some regions and to present survey-robust confidence intervals in visualizations.

NoteNote

This post is sourced from my contribution to El Quantificador, a blog on social issues created by LIDE, an Ecuadorian research nonprofit organization. See the original post in Spanish here.

En Español

El trono en Ponceano: Liga de Quito en cifras con R

En 2023, LDU Quito logró llegar a la final de la Copa Sudamericana tras más de una década sin títulos internacionales. Analizo la trayectoria del equipo en campeonatos internacionales usando R y fuentes como Transfermarkt y FBref.

Homosexualidad: ¿qué piensan los ecuatorianos?

Coauthor: Alonso Quijano-Ruiz. Análisis de datos de opinión pública del World Values Survey sobre las actitudes de los ecuatorianos hacia la homosexualidad. ISSN 2953-6456.

Tres problemas para el análisis cuantitativo de los femicidios en Ecuador

Coauthor: Alejandra Marchán. Análisis de datos administrativos del gobierno ecuatoriano usando R, identificando tres problemas importantes que impiden un diagnóstico claro de la violencia de género. ISSN 2953-6456.

Una perspectiva cuantitativa sobre las protestas en Ecuador

Versión original de Analyzing public opinion from survey data, escrita en español para El Quantificador. ISSN 2953-6456.