Stats SA, Council stand by Census results

Statistics South Africa and the Statistics Council have defended the accuracy of the results of Census 2011, saying they were in line with universally accepted standards and were reached after much consideration.

Following concerns raised on the accuracy of the results and claims that they were rushed by Prof Moultrie and Prof Dorrington – consultants who advised the Council – Stats SA and the Council on Thursday said they stood by the results.

Chair of the Stats Council, Howard Gabriels, said the process of evaluating the latest Census started as far back as 2008, adding that there was extensive and long engagement with Stats SA throughout.

A number of consultants were engaged, including an international team of experts, as the processing stage of the data was nearing finalisation.

Gabriels said the Council considered the report by Moultrie and Dorrington very seriously.

He stressed that he even went as far as stopping the meeting and asked each member of the council whether, based on the report, they believed the council should advise that the Minister in the Presidency for National Planning should not release the results.

Each member said the minister should be advised to release the results and that they were comfortable that the Census was fit for use and could be released, he added.

He refuted “insinuations” that council did not apply its mind to the technical detail and that advice was ignored.

Gabriels pointed out that if was only after much in-depth analysis that the council – made up of respected statisticians, social scientists and economist – took a decision on whether the results being released were consistent, plausible and fit for use.

It was a conclusion that was reached after interrogation of all the technical issues.

He reiterated that the Council did not take the decision lightly and that it had taken its decision, having seriously considered the submissions of Moultrie and Dorrington.

On the issue of whether the results of the Census were rushed or not, Gabriels said after the reports from Moultrie and Dorrington, the council approached the minister to delay the results by a few weeks, from the initial date of 9 October.

The delay was agreed on, giving the council an additional three weeks to work on the data, Gabriels pointed out.

“It is not that the Council rushed; it has taken this decision [well] informed …” Gabriels added.

Statistician-General Pali Lehohla noted that Stats SA abided by the quality standards that were universally accepted. –