Surnames and ancestry in Brazil

Area of Science:

• Sociology
• Computational Linguistics
• Demography

Background:

• Brazilian society is characterized by diverse ethnic origins.
• Accurate classification of ancestry is crucial for socioeconomic analysis.
• Previous methods for surname ancestry classification were limited.

Purpose of the Study:

• To develop and validate a computational method for classifying Brazilian surname ancestry.
• To analyze the socioeconomic impact of surname-based ethnic origins.
• To identify patterns of immigration and settlement in Brazil.

Main Methods:

• Utilized historical records and a database of over 46 million workers.
• Applied fuzzy matching and machine learning classification algorithms.
• Validated results against 1920 Census data and surname geographic distribution.

Main Results:

• Successfully classified the ancestry of 96.7% of single surnames using fuzzy matching.
• Identified five main ancestry categories: Iberian, Italian, Japanese, German, and East European.
• Demonstrated significant associations between surname ancestry and differences in wages and schooling.

Conclusions:

• The developed method provides an accurate and scalable approach to classifying surname ancestry in Brazil.
• Surname ancestry is a significant determinant of socioeconomic outcomes.
• This research offers valuable insights for demographic and historical studies in Brazil.