Although numerous estimates of the Nigerian population were made during the colonial period, the first attempt at a nationwide census was during 1952-53. This attempt yielded a total population figure of 31.6 million within the current boundaries of the country. This census has usually been considered an undercount for a number of reasons: apprehension that the census was related to tax collection political tension at the time in eastern Nigeria logistical difficulties in reaching many remote areas and inadequate training of enumerators in some areas. The extent of undercounting has been estimated at 10 percent or less, although accuracy probably varied among the regions. Despite its difficulties, the 1952-53 census has generally been seen as less problematic than any of its successors.

Subsequent attempts to conduct a reliable postindependence census have been mired in controversy, and only one was officially accepted. The first attempt, in mid-1962, was canceled after much controversy and allegations of overcounting in many areas. A second attempt in 1963, which was officially accepted, also was encumbered with charges of inaccuracy and manipulation for regional and local political purposes. Indeed, the official 1963 figure of 55.6 million as total national population is inconsistent with the census of a decade earlier because it implies a virtually impossible annual growth rate of 5.8 percent. In addition to likely inflation of the aggregate figure, significant intraregional anomalies emerge from a close comparison of the 1953 and 1963 figures. In portions of the southeast, for example, the two sets of data imply that some nonurban local government areas (LGAs) had increased at a rate of almost 13 percent per year, while other neighboring areas experienced a minute growth rate of 0.5 percent per year. Despite the controversy, the results of the 1963 census were eventually accepted.

After the civil war of 1967-70, an attempt was made to hold a census in 1973, but the results were canceled in the face of repeated controversy. No subsequent nationwide census had been held as of 1990, although there have been various attempts to derive population estimates at a state or local level. Most official national population estimates are based on projections from the 1963 census.

The great improvements in transport and accessibility of most areas, in technological capability, and in the level of education throughout the country, as well as the generalized acceptance of national coherence and legitimacy, favored the success of the fall 1991 census. It was to be conducted in about 250,000 enumeration areas by the National Population Commission, with offices in each of the country's LGAs. To reduce possible controversy, religious and ethnic identification would be excluded from the census forms, and verification of state results would be handled by supervisors from outside the state. Some analysts believe that the effort to carry out a reliable census with perceived legitimacy might become an unexpectedly positive exercise, reinforcing a sense of shared nationhood and providing a model for164 or the attempt to overcome regional and ethnic differences.

Data as of June 1991