How do we measure underemployment?

In the United States, official unemployment is determined narrowly; that is, a ratio of persons able and willing to work but do not have a job to the total civilian labor force. The total civilian labor force is defined as the sum of all persons who have a job—any job—plus those who are actively seeking work.  Someone who works only a few hours per week, or works without pay for 15 or more hours in a family-owned business is counted as employed.

The U.S. Census Bureau and the Bureau of Labor Statistics began to realize that the official unemployment rate was not providing a true picture of what was happening in the job market. In 1994, new measures were added to include “discouraged workers” (the individual has a job market-related reason as opposed to a personal reason for not currently looking for work), the “marginally attached” (neither working nor looking for work but indicate they want and are available for work and have looked for work sometime in the past 12 months), and the “involuntary part-time” workers. However, official unemployment is still reported using the traditional definitions and ratios. There is no regular data collection at the national level that measures skills/educational underemployment, with virtually the only attention to this issue being from academics.