Learning About Black Life Through Memorials: The Funeral Programs Project

• In Event: Implications of AI for Qualitative, Quantitative, and Computational Sociological Research

Abstract

Historical restrictions on newspaper publication of the obituaries of Black Americans led Black Americans to develop a unique cultural tradition of detailed funeral programs. Before the advent of personally accessible professional printing services, local Black printers created funeral programs as they created programs for local religious, civic, and fraternal organizations. Today, funeral services widely note the prominence and importance of funeral programs for Black Americans in particular. This project seeks to use funeral programs to expand the dimensions of life course analysis for Black Americans. Funeral programs give an in-depth look into the lives of individuals at a level rarely seen in other data sources. Analyzing and formalizing Black funeral programs and systematizing their data provides a new source of rich biographical and personal information for Black life course research. In addition, funeral program information can be matched to administrative data as a check on the accuracy of information known by loved ones and those recorded in other sources. I describe a project to collect more than 15,000 funeral programs for Black Americans, concentrating on those from the mid- to late 20th century. I describe the process of manually collecting funeral programs and the novel information they contain to describe the new dimensions of research possible by analyzing funeral programs. The large scale collection of funeral programs will construct a unique algorithm using Large Language Models (LLMs) and Natural Language Processing (NLP) to extract data from funeral programs at scale. Unlike existing methods of textual analysis and record linking, this project uses a different source for algorithm training and external validity. The verification check of the similarity between the manual data and the data produced by the algorithm increases the reliability of the method to produce large-scale, novel data on Black life course, and establishes practice for future research.

Author

• Trevon D. Logan, Ohio State University