Poor and often illiterate villager farmers in Central America have been taught by World Neighbors programs to (a) establish and manage simple agricultural experiments with one variable, (b) measure accurately the results, and (c) report the results to the program and each other. The impact on the areas' yields of the technology thus generated, validated, spread, and/or locally adapted, has been significant. Although the situation there differs from that of the United States, and therefore the experimental design would differ, many of the principles and advantages of such a system would probably also apply in North America.