Major Urban Area Studies in Latin America

Latin America covers a vast areal expanse, including a variety of terrain types (the tropical islands of the Caribe, the 8000 km long Andes mountain range, the Amazon rain forest, the windy Tierra del Fuego, the awesome force of the Foz do Iguaçu, the dry Atacama Desert, the volcanoes of Central America and, above all, the man-made wonders of some of the largest cities in the world: Mexico City, Buenos Aires, São Paulo, etc.). For research purposes, it is clearly very difficult to conduct studies which cover the entire area. Therefore, very often, research studies are conducted in several urban areas only.

There are usually several reasons invoked to justify this choice. The primary reason is cost, since it is more expensive and time-consuming to send researchers paddling up the Amazon or hiking across the Andes to conduct audits and surveys. It becomes cheaper and more convenient to conduct the research in the major urban areas, where the research infrastructure is well-developed. A secondary reason is that the subject of the research (such as media usage and product consumption) is supposedly concentrated in the major urban areas, at least for now. Another secondary reason is that the eventual application of these research results will be among the major urban areas anyway. But in so doing, one runs the risk of missing many good opportunities.

Ultimately, the justification for limiting the study to major urban areas depends on how different the results might have been. Since the study included only the major urban areas, it is impossible to tell what the differences might have been. For some variables (such as age/sex distributions), there may be some external benchmark (such as census data). But for other variables, which often represent the principal objectives of the study, the decision reduces down to an act of faith.

In the Los Medios y Mercados de Latinoamérica study, we attempted to cover as much of the population as possible. In the 19 countries in the study, the sample covers over 90% of the population, including all of the major urban areas. The missing 10% are basically too remote or dangerous to reach (such as Patagonia in Argentina, the Norte region in Brazil, etc.). This study provides an opportunity to compare survey results from the major urban areas against a more general population.

For the purposes here, we define 'Major Urban Areas' to be the population which reside in the following thirty Latin American cities: Buenos Aires (Argentina); La Paz (Bolivia); Belo Horizonte, Brasilia, Curitiba, Rio de Janeiro, São Paulo, Salvador (Brazil); Santiago (Chile); Bogotá, Cali, Medellín (Colombia); San José (Costa Rica); Santo Domingo (Dominican Republic); Guayaquil, Quito (Ecuador); San Salvador (El Salvador); Guatemala City (Guatemala); Tegucigalpa, San Pedro Sula (Honduras); Guadalajara, Mexico City, Monterrey (Mexico); Managua (Nicaragua); Panama City (Panama); Asunción (Paraguay); Lima (Peru); San Juan (Puerto Rico); Montevideo (Uruguay); and Caracas (Venezuela).

These cities are the political, economic and cultural centers in their respective countries. Together, these major urban areas account for 28% of the universe in the study. We note that the typical research studies cover much fewer than the 30 listed above. We also note that the remaining 72% of the study universe is not entirely rural; in fact, Latin America is highly urbanized and much of the remaining 72% consists of smaller cities.

The first issue that we will address is the share of socio-economic power, media usage and product consumption that these major urban areas hold. If the major urban areas were identical to the rest of the region, we would expect to see the same 28% across every socio-economic class, media user group and product consumer group.

In terms of socio-economic power, it is standard practice in Latin America to use a definition that is based primarily upon a composite of the presence of certain consumer goods and services. The results are shown in the following table:

Socio-Eocnomic Level % in Major Urban Areas
Level A 42%
Level B 37%
Level C 27%
Level D 19%

(source: Audits & Surveys Worldwide, Los Medios y Mercados de Latinoamérica 1996)

The major urban areas account for more than their share in the upper- and middle-upper classes based upon population size alone. However, it is still true that more than half of the Level A/Level B population live outside these major urban areas.

Media usage is shown in the next table.

Media Usage % in Major Urban Areas
Has Latin-feed Multichannel TV Service 33%
Watched Any Regional Cable/Satellite TV Service last week 34%
Read Any Regional Magazine (at least 1 out of 4 issues) 34%

(source: Audits & Surveys Worldwide, Los Medios y Mercados de Latinoamérica 1996)

The above table might surprise some people, who expect to see multichannel (that is, cable/satellite) television to be concentrated in the major urban areas. The fact is that multichannel television in many countries developed evenly inside and outside the major urban areas. In some instances, it proved physically easier to cable smaller cities. For example, one of the first cable companies in the world to offer cable modem access to the Internet is Cable Aconcagua in Los Andes, a city with a population of 55,000 in Chile.

Here is an assorted selection of some product consumption statistics:

Product Consumption

% in Major Urban Areas
Used toothpaste in last 7 days 28%
Used personal care soap in last 7 days 28%
Used shampoo in last 7 days 29%
Drank carbonated soft drinks in last 7 days 29%
Drank beer in last 7 days 30%
Bought cigarettes in last 7 days 32%
Bought athletic shoes in last 12 months 30%
Bought jeans in last 12 months 30%
Has telephone service 36%
Owns automobile 31%
Owns video game system 38%
Owns electronic organizers 48%
Owns computer at home 47%
Uses computer at work 43%

(source: Audits & Surveys Worldwide, Los Medios y Mercados de Latinoamérica 1996)

These results suggest that, for many basic categories, the major urban areas do not account for considerably more than their share on the basis of population size. The exceptions are certain leading-edge consumer electronics products which have not yet diffused to the rest of the country.

Even if the major urban areas do not account for more than their share on the basis of population size alone, and precisely for that reason, they may serve as the surrogate for the remaining population. But for this to hold, we need to be sure that they share the same detailed behavior beyond just the overall category usage.

For example, we saw that the share of the major urban areas among athletic shoe buyers is 30%, which is about the same as the 28% share of population. However, when we looked at individual brands, we see that the share of the major urban areas among the major brands (such as Adidas, Converse, Fila, Nike, Puma, Reebok, etc.) ranges between 17% and 46%. Upon detailed analysis and research, it became clear that some brands have chosen to concentrate on just a few countries in the region, other brands have focussed on the major urban areas and still other brands have followed a region-wide approach. With this wide range of penetration levels, a major urban area study may yield a quite incorrect portrayal of the situation in the region as a whole.


We acknowledge that most research budgets do not allow for a regionally representative sample. But in conducting a study in just a few major urban areas, one has to be extremely careful about how the results can be validly projected and interpreted.

(posted by Roland Soong on April 24th, 1997)

(Return to Zona Latina's Home Page)