Sleeping soundly

Feelings of safety: based on perceptions or reality

---

Published in Nedbank ISS Crime Index
Volume 5 2001
Number 2, March - April

A national survey conducted in late 2000 asked respondents about their feelings of safety. The survey reveals how feelings of safety are influenced by where people live, their race, age, and wealth. Feelings of safety are often influenced by perceptions about the risk of crime rather than by facts. In many cases, however, individual feelings of safety are closely related to the actual risk of crime.

Declining safety

The Human Sciences Research Council (HSRC) has, for several years, conducted regular national public opinion surveys. Among other things respondents are asked about their feelings of safety, with the specific question: ‘How safe or unsafe do you feel personally?’

Feelings of safety have declined significantly since 1994. In 1994 the majority, or 73%, of respondents said they felt safe or very safe, while 16% felt unsafe or very unsafe. In 2000, 44% of respondents felt safe or very safe while 45% felt unsafe or very unsafe.

The latest HSRC survey, conducted in September 2000, provides some interesting information on how feelings of safety vary according to where people live, their race, age, and standard of living.

Urban and rural

Over half of rural respondents stated that they felt safe (i.e. safe or very safe). Among urban respondents just over 40% felt similarly safe, while just over a third of respondents living in the metropolitan areas stated that they felt safe (Figure 1).

Metropolitan respondents were right in feeling less safe than their urban and rural counterparts. During 1999, 55 murders per 100000 of the population were recorded in South Africa as a whole. With the exception of Pretoria, the rate was higher in the metropolitan areas: 117 in Johannesburg, 82 in Durban, 68 in Cape Town, and 37 in Pretoria.

It is a world-wide phenomenon that certain crimes are more prevalent in cities than in rural areas. Most factors associated with high crime rates are far more characteristic of cities than of small towns. Population density, for example, is thought to be associated with crime in that greater concentrations of people lead to increased conflict, greater levels of stress, and competition for limited resources.

Factors that characterise urbanisation, such as overcrowding, unemployment, and increased consumer demands and expectations are also believed to be associated with high crime rates. High levels of gang activity and the availability of firearms are also mainly evident in urban areas.

Provinces

According to the HSRC survey, residents of the Northern Province felt the safest, with over 60% of respondents stating that they felt safe. Above average feelings of safety were expressed by respondents in the Western Cape, Free State and Northern Cape. Eastern Cape and Gauteng respondents felt the least safe (Figure 2).

Generally, feelings of safety of respondents in the different provinces were supported by recorded crime levels in those provinces. For example, Northern Province respondents felt the most safe and least unsafe. During 1999 (the last full year for which recorded crime figures are available) this province had the lowest per capita rate of 14 out of the 17 most serious and prevalent crimes recorded. Gauteng residents, on the other hand, felt the least safe. During 1999 Gauteng had the highest recorded per capita levels of attempted murder, robbery with aggravating circumstances, common robbery and motor vehicle theft, and the third highest per capita rates of rape, serious assault and murder.

However, it is clear that perceptions seem to influence people’s feelings of safety. The Western Cape is perceived to be a safe province when it comes to crime, but recorded crime figures show a different picture. During 1999 the province had the highest per capita rate of recorded murder, burglary and common assault, and the second highest rate of attempted murder, rape, serious assault, theft of motor vehicles and common robbery. Despite this, respondents in the Western Cape felt the most safe after Northern Province respondents.

Race

When broken down into racial groups, coloured respondents felt the safest, followed by black, white and Asian respondents. Asian respondents also felt the most unsafe by a wide margin (Figure 3).

The relatively high Asian and white feelings of insecurity do not seem to be related to their vulnerability. According to the National Victims of Crime Survey, 7% of blacks and coloureds were victims of violent individual crimes such as assault, sexual offences, and robbery, followed by whites (5%), and Asians (3%). ISS victim surveys support the finding that blacks and coloureds are disproportionately likely to be victims of violent crime, while Asians and whites are more likely to fall victim to property crimes. (See Crime Index Volume 2, Number 1, ‘Crime and policing: Perceptions and fears’.)

Age

Respondents in the age group 25 to 34 felt the safest, followed by those over the age of 50. Those aged 35 to 50 years felt the most unsafe of the four age categories surveyed (Figure 4).

There appears to be no direct correlation between people’s feelings of safety and their risk of victimisation. Young people are disproportionately likely to be victims of crime, especially young urban males. The National Victims of Crime Survey found that of all individuals who had experienced at least one violent crime in South Africa in 1997, almost a third were aged 16 to 25. The level of victimisation consistently declines with age thereafter. For non-violent crime, the age group 26 to 35 years experienced the highest level of victimisation, after which the level of victimisation declines rapidly with age.

Lifestyle

The HSRC survey measured feelings of safety according to respondents’ living standards measures (LSM). LSM categorises individuals according to an eight-point scale based on the possession of a range of durable household articles, access to reticulated water and electricity supplies, and a range of other income-related variables. 8 is the highest LSM category, 1 the lowest.

With the exception of the highest LSM category, the lower down on the LSM scale respondents were, the safer they felt. Among the poorest of the poor — LSM 1 — some 61% of respondents stated that they felt safe, while only 35% of respondents belonging to LSM 7 said the same (Figure 5).

According to the National Victims of Crime Survey the wealthiest households are at greatest risk of falling victim to household-related property crime. In respect of violent household-related crime the correlation is weaker. The wealthiest households experience less violent crime than middle-class households, but more than very poor households. Individuals living in the wealthiest households are the most likely victims of individual property crimes but the least likely victims of violent crimes. The opposite is the case for individuals living in the poorest households.

Conclusion

People’s feelings of safety are influenced as much by actual recorded crime levels as by perceptions. This is an important point for criminal justice policy makers. Even an efficient criminal justice system can hope to make only a limited difference to levels of crime. The state can, however, reassure those most fearful of crime — the wealthy and the old — that they are unnecessarily fearful of many types of violent crime.

Martin Schönteich
Institute for Security Studies