Options after census results quashed

Government experts are still undecided on how to sort out the problem of the cancelled population count for eight districts in northern Kenya.

Three options are on the table but no agreement has been reached on which to use nearly a month after the cancellation.

The results were nullified after they showed an abnormal rise in population — of about a million.

Top of the list of the options is a straight head count of all residents in the affected areas.

This is viewed as one way to help provide quick and accurate data for planning as it may take as short as a day to complete.

In this case, varied data including the number of houses, animals and occupations might not be necessary.

The argument is that the rest of the data collected during the census will provide additional information to complement the head count.

That would mean the State will uphold results of data like levels of education, number of deaths and births.

Also being floated is a complete census of all the items of last year’s questionnaire.

In this case, the government will collect and process data in much the same way it did during the national census last year.

It is not clear, however, whether there will be another pilot census or a post-enumeration exercise that usually accompany such counts.

The third option would require that the government uphold the faulty results but population experts will then use scientific means to adjust them to remain relevant for use over the next 10 years.

The results were cancelled in Mandera Central, Mandera East, Mandera West, Lagdera and Wajir East. Other areas were Turkana Central, Turkana South and Turkana North.

Dr Collins Opiyo, the director of population and social statistics, said the experts were yet to make their proposals known to Planning minister Wycliffe Oparanya, who announced the cancellation.

Dr Opiyo told the Nation that a repeat of the census would produce the most comprehensive results but its costs would be “huge”.

“We are dealing with a big problem here,” he said.

The country’s top demographer said the bureau had been asked to put a proposal to sort out the problem.

“We will come up with details on what (correcting the anomalies) will entail, the budget and other logistics,” he said. “We will decide whether to go for a repeat or head count.”

Mr Oparanya announced the cancellation when he released the census results on August 31, saying the figures were heavily exaggerated.

Remedial measures taken

His announcement seemed to vary from his remarks in a foreword he did in one of the census volumes, which did not mention repeating or cancellation of the results.

In the foreword, Mr Oparanya said “the data will be assessed further using internationally acceptable procedures and suitable remedial measures undertaken”.

He said the data from the affected districts were “anomalous” and “exhibit population growth rates that deviate significantly from the patterns portrayed not only by the rest of the country but by their respective neighbouring districts as well”.

Mr Oparanya advised the public and those interested in using the data of the affected districts to take caution. But leaders from the areas have been demanding that the government rescind its decision to nullify the results.

Dr Opiyo explained that a repeat census will need massive resources, a budget the Kenya National Bureau of Statistics did not immediately have.

But the government can easily obtain the funds if the option of repeating the census was accepted, he added.

He supported Mr Oparanya for cancelling the results since they showed a big deviation from the projected results.

They were also much different from the results of the pilot phase of the census and a mini-census conducted after the official count.

The post-enumeration exercise checks the accuracy of results obtained from the main census.

“Our post-count survey, conducted in December, shows a large deviation from the official count in the affected areas,” said Dr Opiyo.

“This suggests that the main census may have over-enumerated people.”

The population of the affected areas had been projected to stand at about 1.4 million people. But the actual census put out a result of 2.3 million people, a figure Dr Opiyo termed “abnormal”.