Latin Americanists commonly stress the traditional cultural and philosophical differences between the region they study and the United States. A frequent contention holds that such historical contrasts persist to the present. For example, Howard J. Wiarda asserts, “Latin America … remains paternalistic, hierarchical, authoritarian, Catholic, corporate, personalist, and elitist to its core.” In contrast, the United States is presumably more egalitarian, Protestant, and impersonal than her southern-hemispheric neighbors.

This report on the special sessions of Latin American historical demography, at the XLI International Congress of Americanists in Mexico City (2-7 September 1974), was prepared to present some current trends and research methods in the field. The sessions, held at the National Museum of Anthropology on 5-6 September, were preceded by an informal meeting of the participants at which the presentation and discussion of papers was organized. Elsa Malvido, coordinator of the special sessions, opened it with a few brief words of welcome.

In recent years, a considerable effort has been made in political science to facilitate cross-national research of a truly comparative nature. One common strategy is to assemble statistical data from as many nations as possible that then can serve as quantitative indicators for a variety of political, social, and economic phenomena. The use of quantitative operationalizations allows for statistical testing of appropriate hypotheses. While the goals of such data collection are most certainly praiseworthy, a variety of questions can and ought to be raised about the quality of much of the data made so readily available to today's researchers. My purpose here is to illustrate the risks involved in the unwary use of one such “quantitative indicator”—trade union electoral statistics. What is true for the election data is also true in varying degrees for other types of statistical information from Latin America.