For many years, the U.S. measured literacy simply by asking respondents whether they could read or write, an approach perhaps sufficiently adequate when most people worked with their hands. Using this method, the Census Bureau in 1969 estimated that illiteracy in the U.S. population age 14 and older was only 1 percent.

In 1992 the U.S. Department of Education embarked on a more thorough analysis and mounted a landmark survey. It asked a representative group of 26,091 Americans to read several texts and then had them demonstrate that they understood the texts. The department used three types--prose, document and quantitative texts--containing fairly simple material encountered in everyday life. The study was repeated in 2003.