Literacy in the United States

There is an article in the current Scientific American (September 2006) by Rodger Doyle on the results of a study of literacy in the United States. It divides literacy into three types:
· Prose Literacy (The skills needed to comprehend and use information from continuous texts such as newspaper articles.)
· Document Literacy (The knowledge and skills needed to comprehend and use information from noncontinuous texts such as simple statistical tables.)
· Quantitative Literacy (The knowledge and skills needed to identify and perform computations using numbers embedded in printed materials such as tax forms. Examples: calculating a tip or loan rate.

The study showed that prose and document literacy were decreasing slightly for the total population, but that quantitative literacy was increasing, based on the difference in performance measured in the survey in 1992 and 2003. However the results were worse for high school graduates, college graduates and those with graduate degrees than for the population as a whole. Not only did prose and document literacy decrease notably for all three groups, but quantitative literacy decreased quite a bit for college graduates and those with graduate degrees. Sad news/

It occurred to me to post on the interpretation of this data. How can overall changes in literacy look better than the changes in literacy for high school, college and graduate school graduates? It is possible, for example, that the remaining population (e.g. non high school graduates) did much better on the tests in 2003 than in 1992.

Perhaps more likely that people were getting more education in 2003 than in 1992. Since it is likely that high school grads are more literate than those who don’t graduate from high school, that college grads are more literate than high school grads who don’t graduate from college, and that people with graduate degrees are more literate than college grads without graduate degrees, increasing education levels in the population might account for performance of the total population appearing better than performance for any of the subpopulations with specific levels of education.

Indeed, if the cohorts of people completing each level of education are increasing as a percentage of the overall population, I would expect literacy levels to decrease at each level of education. Those who would have dropped out of school in the past because they were having trouble, would be more likely to stay involved;

The article references two studies:

· A First Look at the Literacy of America’s Adults in the 21st Century
· The Twin Challenges of Mediocrity and Inequality: Literacy in the U.S. from an International Perspective