OPINION

Did the SAPS bungle their crime stats?

Changes in the number of reported crimes and changes in the crime rate don't reconcile

On Thursday Police Minister Nathi Mthethwa presented the SAPS 2012/2013 national crime statistics. In his statement he listed the increases or decreases of various categories of crime in terms, not in absolute numbers, but in the rate or ratio per 100 000 of the population.

Thus he stated, for instance, that total sexual offences decreased by 0.4% and robbery in non-residential premises by 0.6%. He added that there had been an increase of 1.2% in aggravated robbery, 3.3% in residential burglary, 0.6% in commercial crime, and 3.6% in residential robberies.Mthethwa also stated that "we have witnessed a slight increase of 0.6%" in the murder rate.

In absolute terms however there were 16 259 murders in 2012/2013, 620 up from the 15 609 recorded in 2011/2012. This is a 4.2% increase. How then can the differential between a 0.6% increase in the murder rate be reconciled with this 4.2% increase in the total number of murders?

The short answer is, it can't.

One possible explanation is that the population of South Africa grew by 3.3% between 2011 and 2012. However, Statistics South Africa estimates that the population of South Africa only grew by 1.34% between 2011 and 2012 (see here.)

In fact, what seems to have happened is the following: The 2011/2012 crime statistics' ratios per 100,000 of the population were calculated using Stats SA's 2011 mid-year population estimate of 50 586 757. For the 2012/2013 crime statistics the SAPS worked out the ratios using the Stats SA estimate of a population of 52 274 945 in 2012. It then worked out the change in the ratios per 100 000 of the population between the two.

However, the differential between the 2011 StatsSA estimate used by the SAPS and the 2012 one is 1 688 188, an implausibly large 3.34% gap. This doesn't reflect a real change in the population, but rather an adjustment upwards in StatsSA's estimate of the population (presumably post-Census 2011). According to these adjusted estimates the population in 2011 was 51 579 548 - only 695 397 (or 1.35%) fewer than in 2012.

If one uses the updated population figures of StatsSA for 2012 and 2011 then the murder rate actually increased by 2.78% (not 0.6%). There was a 1.53% increase in the rate of sexual offences (not a decrease of 0.4%) and an increase of 1.30% in non-residential robberies (not a decrease of 0.6%). The rate of aggravated robbery was up 3.24% (not 1.2%), that of residential burglary by 5.33% (not 3.3%), commercial crime by 2.61% (not 0.6%), and residential robbery by 5.65% (not 3.6%).

The table below sets out the change in the number of crimes committed, in various categories, between 2011/12 and 2012/2013 - and then the change in the ratio per 100 000 of the population using the old Stats SA population estimates for 2011 versus the new ones. 

Table 1: Changes in the crime rate in 2012/2013 using the old StatSA 2011 population estimates (used by the SAPS) vs changes in crime rate using new StatsSA 2011 population estimates:

Crime Category

2011/ 2012

2012/ 2013

Change in absolute numbers

% change in absolute numbers

Change in rate using old 2011 population estimate of 50 586 757

Change in rate using new 2011 population  estimate of 51 579 548

CONTACT CRIME (CRIMES AGAINST THE PERSON)

Murder

15 609

16 259

650

4.2%

0.80%

2.78%

Total Sexual Crimes

64 514

66 387

1 873

2.9%

-0.42%

1.53%

Attempted murder

14 859

16 363

1 504

10.1%

6.57%

8.66%

Assault with the intent to inflict grievous bodily harm

192 651

185 893

-6 758

-3.5%

-6.62%

-4.79%

Common assault

181 670

172 909

-8 761

-4.8%

-7.90%

-6.09%

Common robbery

52 980

53 540

560

1.1%

-2.21%

-0.29%

Robbery with aggravating circumstances

101 203

105 888

4 685

4.6%

1.25%

3.24%

CONTACT-RELATED CRIMES

Arson

6 418

6 064

-354

-5.5%

-8.57%

-6.77%

Malicious damage to property

122 221

121 113

-1 108

-0.9%

-4.11%

-2.22%

PROPERTY-RELATED CRIMES

Burglary at non-residential premises

70 041

73 630

3 589

5.1%

1.73%

3.73%

Burglary at residential premises

245 531

262 113

16 582

6.8%

3.31%

5.33%

Theft of motor vehicle and motorcycle

59 097

58 370

-727

-1.2%

-4.42%

-2.54%

Theft out of or from motor vehicle

130 475

139 658

9 183

7.0%

3.58%

5.61%

Stock-theft

30 949

29 894

-1 055

-3.4%

-6.53%

-4.69%

CRIMES HEAVILY DEPENDENT ON POLICE ACTION FOR DETECTION

Illegal possession of firearms and ammunition

14 461

14 872

411

2.8%

-0.48%

1.47%

Drug-related crime

176 307

206 825

30 518

17.3%

13.52%

15.75%

Driving under the influence of alcohol or drugs

69 441

71 065

1 624

2.3%

-0.97%

0.98%

OTHER SERIOUS CRIMES

All theft not mentioned elsewhere

377 432

362 816

-14 616

-3.9%

-6.98%

-5.15%

Commercial crime

88 050

91 569

3 519

4.0%

0.64%

2.61%

Shoplifting

71 848

71 309

-539

-0.8%

-3.96%

-2.07%

SUBCATEGORIES FORMING PART OF AGGRAVATED ROBBERY ABOVE

Carjacking

9 475

9 990

515

5.4%

2.03%

4.03%

Truck hijacking

821

943

122

14.9%

11.15%

13.33%

Robbery at residential premises

16 766

17 950

1 184

7.1%

3.60%

5.64%

Robbery at non-residential premises

15 951

16 377

426

2.7%

-0.65%

1.30%

OTHER CRIME CATEGORIES

Culpable homicide

11 788

11 395

-393

-3.3%

-6.46%

-4.62%

Public violence

1 152

1 783

631

54.8%

49.78%

52.72%

Crimen injuria

31 858

29 040

-2 818

-8.8%

-11.79%

-10.06%

Neglect and ill-treatment of children

2 949

2 758

-191

-6.5%

-9.50%

-7.72%

Kidnapping

3 874

4 333

459

11.8%

8.24%

10.36%

Click here to sign up to receive our free daily headline email newsletter