The June increase in the unemployment rate was the lowest in eight months, but economists said that number doesn't fully reflect the impact of the recession on U.S. workers.

The Federal Bureau of Labor Statistics reported yesterday that the seasonally adjusted unemployment rate edged up to 9.5 percent -- a 26-year high -- with 218,000 more people in the country jobless in June than in May, when unemployment hit 9.4 percent.

However, payroll employment -- a different survey that counts the number of jobs, both full- and part-time, and not the number of people in them -- fell by 467,000, far more than the 322,000 jobs lost in May.

Someone who had two part-time jobs and lost one would not be considered unemployed but the lost job would be counted in the payroll employment numbers.

Overall, the number of U.S. jobs, full- and part-time, is down 6.5 million since the recession began in December 2007.

In addition, lost in the broad national numbers are the extent of the hits taken by undercounted groups such as Hispanics and African-Americans.

Despite the smaller increase of jobless people in June, there is evidence in the bureau's report that those who were out of work are having a difficult time finding jobs.

The number of people who have been unemployed for 27 weeks or more -- the so-called long-term unemployed -- increased by 433,000 in June to nearly 4.4 million.

Harry J. Holzer, an economist with Georgetown University and the Urban Institute in Washington, D.C., said other numbers in the report more accurately reflect what is happening in the labor force than the widely reported unemployment rate.

Mr. Holzer said a better indicator of the country's overall economic health is a table put out by the Bureau of Labor Statistics that reports in June 16.5 percent of people who would work if they could are either unemployed, marginally attached (defined as people without jobs who have given up looking for work) or stuck in part-time jobs and can't get enough hours to be full time.

That number has been steadily climbing.

High unemployment in certain groups also may be underreported.

Mr. Holzer said Hispanics had an unemployment rate of 11.8 percent.

Yet that group is heavily represented in the home construction labor force, which has been hit hard in this recession. Because undocumented construction workers are unlikely to answer a government survey, the rate may not fully reflect the unemployment.

Black men, too, actually have higher rates than the official unemployment rate of 16.4 percent reported in June, Mr. Holzer said.

Teenagers, officially those aged 16 to 19, had particularly high rates of unemployment. Unemployment for white teens last month was reported at 21.4 percent while black teens were far worse off with a 37.9 percent unemployment rate. That translated to 1.16 million unemployed white teenagers and 276,000 unemployed black teenagers.

Overall unemployment officially hit 14.7 percent for black men and women of all ages, but Algernon Austin, director of the Program on Race, Ethnicity and the Economy for the Economic Policy Institute in Washington, D.C., said it is probably closer to 16.5 percent. It's difficult to contact and accurately survey unemployed people.

He said African-American unemployment would have climbed higher if that population had been swept into the housing boom that led up to the housing bust.

Mr. Austin said the silver lining, such as it is, is that in most recessions, the official overall unemployment figure for blacks is normally twice that of the national unemployment rate. Currently the unemployment rate for black people is only 1.75 times higher.