The nation's unemployment rate was unchanged in May, holding at 6.3 percent for the second straight month as the economy absorbed most of a whopping 658,000-person increase in the labor force, the Labor Department announced yesterday.

The 6.3 percent jobless rate is the lowest since April 1980, and is 4.5 percentage points lower than the peak rate of the last recession, 10.8 percent in November 1982. In recent months, however, the monthly improvements in the job situation have been relatively small -- the rate has declined by less than one percentage point in the last year.

The number of civilians with jobs rose about 612,000 last month, and the ranks of the jobless swelled by 46,000, the Bureau of Labor Statistics said. Most of the new job holders were adult women.

"Even though employment normally increases strongly between April and May as outdoor activity picks up and young people enter the labor market, the size of the April-to-May {employment} change is probably somewhat exaggerated," BLS Commissioner Janet L. Norwood told the congressional Joint Economic Committee.

She added most of the job growth this year was in the service sector, while manufacturing industries have shown little increase in employment. In May, employment was up in such industries as retail trade, services and finance, insurance and real estate, and down in construction, electronic equipment manufacturing and primary metals. In the last 12 months, only 43,000 goods-producing jobs have been created.

"Manufacturing employment is just about where it was a year ago, while industrial production is up 1.3 percent," said economic forecaster Joel Popkin. "All of the gain in manufacturing output this year was covered by productivity; there was no need to employ more workers."

The apparent slowdown in the American job-creating machine also was underlined by the BLS's annual revisions in one of the two surveys it relies on for employment information. The unemployment rate and aggregate employment figures are reached by direct surveys of households, but the labor statistics bureau also collects employment information from various industries.

Statistical revisions in the industry survey reduced total nonfarm employment by 700,000 jobs in February, the last month for which statistics were computed with and without the revisions.

After the revisions, nonagricultural employment in May as measured by the business survey was 101.7 million, an increase from April of 123,000. The household survey indicated an employment level of 112.4 million. There were 7.5 million Americans looking for work in May, the BLS said.

Economists cite a number of factors to explain the traditional discrepancy between the two surveys, including the fact that farm employment and self-employment is measured in the household survey but not in the business survey. One explanation for the large increase in the labor force in May, economists suggested, was the implementation May 5 of the amnesty provisions of last year's immigration law, under which undocumented aliens were allowed to apply for legal residency. Those applications may have induced workers who had been part of the underground economy to go on their employers' books openly.

Employment among Hispanics rose 137,000 in May, a disproportionate increase compared with that group's representation in the general population.

The static unemployment rate and the seeming exaggeration of the employment increase were more indications of a sluggish economy, according to some economists.

"The numbers suggest that the economy is basically moving sideways," said Popkin.

The BLS reported that the jobless rate for adult men was 5.5 percent, unchanged from April. The rate for adult women was 5.4 percent, down from 5.5 percent last month.

Teen-age unemployment was 17.7 percent, up from 17.4 percent in April, while joblessness among whites was 5.3 percent, down from 5.4 percent last month; for blacks, 13.8 percent, up from 13.0 in April, and among Hispanics, 8.7 percent, down from 9.2 percent. The unemployment rate for black teen-agers, typically a high-unemployment group, was 37.6 percent, up from 36.5 percent last month.

Another measure of unemployment, including members of the armed forces stationed in the United States, also was unchanged last month at 6.2 percent.