Glaring errors in crime stats

The 2012-2013 crime statistics are wrong and misleading, and national police commissioner Riah Phiyega's reassurances about the police winning the war against crime are false.

In fact, security experts say, murder, attempted murder and sexual offences have increased more than the police claimed in September.

Yesterday, the Institute for Security Studies revealed that there were several major errors in the stats - errors that meant that "South Africans don't have an accurate picture of the threats they are facing".

Gareth Newham, head of the institute's crime and justice programme, said the errors in the crime stats were glaring.

"For the first time since 1994, the crime ratios released by the SAPS can be proved to be statistically incorrect," said Newham.

"There has been evidence of statistics being manipulated at some police stations, but not enough to have a [significant] effect on national crime figures."

He said this was the first time that errors in the compilation of statistics had been detected at police national head office level.

However, police yesterday denied the claim saying their stats were correct.

The Institute for Security Studies called for an independent inquiry into how it came about that the crime trends had been miscalculated.

"The effect of this miscalculation is that it downplays the extent to which certain serious violent crime categories are rising, while exaggerating those categories that are decreasing.

"More people were murdered and attacked in their home, business and on the streets than in the previous year.

"More houses and businesses were broken into and burgled," said Newham.

The errors include:

• Police claim that the number of murders and attempted murders nationally increased by 0.6% and 6.5% respectively. The institute's calculations showed a 2.6% and 8.7% increase respectively;
• Police said sexual crimes decreased by 0.4%. The institute calculated a 1.5% increase;
• In Western Cape, police claimed murder and attempted murder rose by 0.5% and 26.4% respectively. The institute said murder increased by 10.1% and attempted murder 38.3%;
• Police said house burglaries in the Western Cape dropped by 0.4%, but the institute said they increased by 9.1%; and
• In Gauteng, police stats show attempted murder dropped by 5.5%; the institute says it rose by 1.8%.

"The errors are the result of inaccurate population figures being used to compare the change in crime ratios between 2011-2012 and 2012-2013," Newham said.

He said the errors resulted in crime trend reporting being "wrong in every category".

"Why did this error occur and at what level was it sanctioned?" he asked.

He said the police used Stat s SA population estimates .

Based on the 2001 census, Stats SA estimated that the population would be 50.6million by 2011. But 2011 census data showed it was 52million.

Police spokesman Lieutenant-General Solomon Makgale said the institute was deliberately sowing confusion. He did not suggest a motive.

"It was argued that crime ratios are crucial and the population estimates we used for the 2011-2012 crime ratios are incorrect. This is not true.

"The raw data, which are actual cases reported, is not based on population figures.

"Therefore one cannot conclude that the overall crime statistics should not be trusted.

"This only serves to confuse the public." Makgale said the police would begin to apply the new population estimates, based on the 2011 census, from the 2012-2013 financial year.

Retired police statistician Chris de Kock agreed with the institute: "If I were still there, I would have rectified the mistake. The police reaction is 'We have our own way of doing this'."

University of Cape Town demographer Tom Moultrie said: "The ISS points are legitimate. The police's data seem to be wrong."