Local officials preparing for the 2010 census are worried about missing individuals in the decennial head count, which could give the region less funding and influence than it deserves.

A research report released yesterday, however, suggests the opposite occurred in Pittsburgh in 2000.

So many Pittsburghers were mistakenly counted twice that the city's reported population of 334,563 was inflated by 752 residents, according to Temple University professor Eugene Ericksen. It was the only one of 11 cities he examined in which there was a census overcount instead of undercount.

His findings were part of a report issued by the Pew Charitable Trusts' Philadelphia Research Initiative, aimed at promoting census participation in Pennsylvania's largest city. Dr. Ericksen, a specialist in census evaluations, estimated Philadelphia's net undercount in 2000 at 8,326 residents, or 0.5 percent of its population.

The census count to be conducted in April is used in allocating more than $400 billion in federal aid annually among states and municipalities as well as for drawing up legislative district boundaries. The house-to-house count is also a baseline for annual population estimates done over the nine succeeding years.

Using a strategy adopted in many urban areas, Mayor Luke Ravenstahl and Allegheny County Executive Dan Onorato formed the City-County Complete Count Committee in July to work with the U.S. Census Bureau and community organizations in promoting complete participation in the count.

If everything occurs as in 2000, that apparently will be no problem. Nothing is being taken for granted, however, said a city official working with the committee.

"The message from both the mayor and county executive makes it clear our job is to do everything we can to make sure everyone's counted and increase our rates from 2000," said Kim Graziani, the mayor's director of neighborhood initiatives.

She was referring to Pittsburghers' voluntary census participation of 61 percent in 2000, meaning those who return mail questionnaires without anyone needing to knock on their door. That was lower than the national average of 67 percent. The city's final count was determined through follow-up efforts by census field workers.

The problem of undercounts among minorities, immigrants and other hard-to-count groups are discussed at length after every census. But if more people are counted twice than are missed, it results in a net overcount. A followup analysis by the Census Bureau of its work in 2000 found there was a national overcount for the first time ever, by 1.3 million residents.

Such overcounts are rare in large cities. Dr. Ericksen found net undercounts in Atlanta, Baltimore, Boston, Chicago, Detroit, Houston, Los Angeles, New York and Phoenix, in addition to Philadelphia.

Pittsburgh has a lower percentage of immigrants and minorities than those cities. It also has a higher share of college students and retirees than most cities, and those are two groups prone to overcounting because they may have two different residences.

Phil Sparks, co-director of The Census Project, a nonprofit group evaluating preparations for the 2010 census, said large-scale community outreach efforts in 2000 led to more people participating than in the past. The bureau had multiple ways in which they could be reported, and some were counted twice without realizing it.

"The Census Bureau would rather have two forms and try to unduplicate them, than to have no forms and have somebody uncounted," Mr. Sparks said. "You can imagine that the computer power needed to unduplicate all of these is formidable, and it's usually successful in 98 percent or more of the cases, but other times it's not."

Because of the federal holiday, no one from the Census Bureau was available yesterday to address the issue, but Mr. Sparks said the overcount would have been too small to give Pittsburgh much unfair advantage.

"What it shows is the last time around you had a good community-inspired effort to get everyone counted," he said. "I would say they must have worked like heck to make sure everyone gets counted -- the last 10 percent is like pulling teeth."

Some cities are using public funding to support their count initiatives this year, but that is not the case in either Pittsburgh or Philadelphia, according to the new Pew report. It cautioned that a complete and accurate national count this year could be made harder by the lack of local funds to aid publicity, compared to a decade ago.

Ms. Graziani said the city and county, for now, are relying on a grass-roots system in which 24 individuals from a cross-section of organizations will head six subcommittees promoting participation. Those individuals will receive special Census Bureau training this month before reaching out to businesses, unions, faith-based groups and educational and community service organizations, she said.

Looking for more from the Post-Gazette? Join PG+, our members-only web site. You'll get exclusive sports content, opinion, financial information, discounts from retailers and restaurants, and more. Our introduction to PG+ gives you all the details.