ALL over Africa, customs officials, border guards, police officers and countless bureaucrats can be seen jotting down information helter-skelter, often on random sheets of paper if there is paper, and if the ink hasn't dried in their pens. At hospitals that have almost no supplies, where patients must provide everything from medicine to surgical tools to be treated, the illnesses of the sick or dead go unrecorded.

Yet there is never a shortage of statistics about the continent: millions of dead from the war in Congo, hundreds of thousands of refugees in West Africa, where a third of the population is H.I.V.-positive. Despite their apparent precision, the numbers are often estimates that can vary according to politics. Figures for AIDS would surely be more precise if its spread had not been denied by some African leaders. To policy makers, humanitarian workers or journalists working in sub-Saharan Africa, one of the hardest things to find is a reliable number. Lack of money and expertise, the collapse of roads and railways that has cut off huge swaths of the continent, all make compiling solid statistics nearly impossible. In many countries, very little is known, statistically speaking, outside the capitals. The latest statistics, or the only ones, are sometimes decades old, from colonial days.

Everyone agrees that Africa's problems are enormous. Whether a particular conflict's refugees really number 400,000, or half that, amounts to quibbling over two morally unacceptable alternatives. But in the real world of limited resources, where Africa is getting less and less attention, do loose numbers do more harm than good? If the experts can't give a real picture, will those numbers breed cynicism about the problems here and the usefulness of aid?

Consider Nigeria. Everyone agrees it is Africa's most populous nation. But what is its population? The United Nations says 114 million; the State Department, 120 million. The World Bank says 126.9 million, while the Central Intelligence Agency puts it at 126,635,626. Nigeria's government's last estimate, a decade ago, was 105 million. The population of Texas -- less than the difference between these last two estimates -- may or may not be living in Nigeria.

Because of the scarcity of numbers here, those that do exist tend to be more politicized and less scrutinized than they are elsewhere. That figures for refugees in a particular war, or victims from a certain illness, are vastly inflated is an open secret. So what? humanitarian officials argue, privately. With Africa stuck on the world's back burner, it is difficult to draw attention without generous statistics. The cause is good.

"There's always a large mixture of scientific accuracy, political imperative and fund-raising," Dr. Ronald Waldman, director of the program on forced migration and health at Columbia University's Mailman School of Public Health, said of statistics in Africa.

For example, he said, health programs tend to be directed at a single cause of mortality. So although a child in Africa or other developing areas usually dies of a combination of illnesses, he said, a program against diarrhea and another against pneumonia might each claim the same child in its statistics, "trying to make itself the most important, in order to attract donor funding."

"In real life, though, kids tend to die of more than one condition at a time," Dr. Waldman said in an e-mail message, adding, "But each program would 'take credit' for the death, so the total number of annual deaths, when you added up the claims of programs, would exceed the actual number of annual child deaths."

The difficulties in arriving at solid numbers in Africa can be measured in the death toll from Congo's four-year war. The commonly cited figure of 2.5 million war dead stems from studies conducted in eastern Congo by the International Rescue Committee, a refugee agency based in New York that operates in eastern Congo.

Les Roberts, an epidemiologist at the agency, led a team that randomly surveyed 2,600 households in seven areas in eastern Congo with a population of 1.5 million according to government statistics from 1996. The researchers compared the mortality rate today with the mortality rate before the war, in 1998. They determined how many deaths -- deaths of people who could no longer get treatment for malaria, deaths of malnourished children, violent deaths -- were caused by the war. The researchers then extrapolated the findings onto all of eastern Congo, a region with 20 million people.

"Extrapolating from 1.5 million to 20 million, it's shoddy, but it's the best we can do right now," Mr. Roberts said. "People correctly criticized us for that."

The agency's figures have been well accepted.

"The methods used in eastern Congo are similar to those used by other organizations in previous conflicts and crises in Africa," said Dr. Bradley Woodruff, an epidemiologist at the Centers for Disease Control and Prevention in Atlanta.

"In Western countries, we have death certificates for each person, which you can just count up," Dr. Woodruff said. "In places where there is no recording of death, you can only estimate."

Estimates can vary significantly by slightly changing a variable. For example, the International Rescue Committee's report of 2.5 million deaths says that by altering certain assumptions, the figure could range from 2.1 million to 3.1 million.

SOMETIMES, however, figures are based not on scientific estimates, but on pure guesswork. In Africa, in the absence of any figures at all, imaginary ones take on a life of their own -- as they did last year with the charges that child workers were forced to work in Ivory Coast's cocoa plantations.

Many accounts in the British and American news media last year spoke breathlessly of 15,000 child slaves on Ivory Coast's cocoa plantations, producing the chocolate you eat.

The number first appeared in Malian newspapers, citing the Unicef office in Mali. But Unicef's Mali office had never researched the issue of forced child laborers in Ivory Coast. The Unicef office in Ivory Coast, which had, concluded that it was impossible to determine the number.

Still, repeated often enough, the number was gladly accepted by some private organizations, globalization opponents seeking a fight with Nestle and Hershey, and some journalists.

Some reports incorrectly cited the State Department's annual human rights report on Ivory Coast as the source of the 15,000 figure. In fact, the State Department report for 2000 said simply that "according to a Unicef study, approximately 15,000 Malian children were trafficked and sold into indentured servitude on Ivorian plantations in 1999." The report for 2001 said that "the number is difficult to estimate" because no "thorough survey has been conducted."

This month, the results of the first extensive survey of child labor in cocoa plantations in Ivory Coast and three other African nations were released by the International Institute of Tropical Agriculture, a nonprofit, multinational research organization that works in Africa. The survey, financed by the Agency for International Development and the United States Labor Department, found that almost all children working in cocoa fields were children of the plantation owners, not forced laborers.

As for child workers unrelated to the plantation owners, the study found that brokers had placed 2,100 foreign children, most of them ages 15 to 17, in Ivory Coast's cocoa plantations. Ninety-four percent of the children, the study says, knew the intermediary, or broker who hired them for the plantation work.

"The most frequent reason given for agreeing to leave with the intermediary was the promise of a better life," the report says. It adds: "None reported being forced against their will to leave their home abode. One hundred percent indicated that they had been informed in advance that they were going to work on cocoa farms."

Jim Gockowski, an American agricultural economist who led the study for the Institute of Tropical Agriculture, said, "By and large, the cocoa industry didn't deserve the rap it got."

Mr. Gockowski, who is based in Cameroon and has worked in African agriculture for a decade, added: "Anyone that's lived in Africa knows kids help out on the farms, probably more in developing countries than developed ones. But even in the United States -- my own background is a farming background -- we grew up helping on the farm. Everyone was pretty surprised when all the wild figures -- 15,000 trafficked children -- were being thrown around."

But politics is sometimes more influential than precision when it comes to numbers in Africa. Since they were released early this month, the institute's findings have received little attention -- perhaps only 1 percent of what the 15,000 figure received.