Trends

Over the four years 1995 to 1999, the number of violent incident measured by the BCS fell by 20%. There was a fall for all violent categories, except robbery and snatch theft.

Notes:

1. Source 1996, 1998 and 2000 BCS.
2. Single asterisks indicate the change is statistically significant at the 10% level (i.e., there is one in ten chance that the observed difference could have arisen by chance).  Double asterisks indicate a change is significant at the 5% level (i.e., one in 20 chance).  Significance has been judged by comparing estimated rates per head of population.

• Robbery increased by 14% and theft from the person by 4%. However, neither of these increases was statistically significant. The increase in robbery is influenced by a marked rise in incidents against 16-year-olds.

Click here for a full size version of the graph

• According to the BCS typology of violent crime, there was a 29% increase in stranger violence but a 19% decrease in acquaintance violence. Mugging and domestic violence showed no statistically significant change.

  For offences that can be compared, the BCS gives a more favourable picture of trends between 1997 and 1999 than do police figures, with larger decreases, or smaller rises (Figure C). The conclusion that BCS figures are more favourable than those of the police is statistically most robust for vehicle thefts, burglary and bicycle theft. The fact that police have more crimes to record because of increased reporting is a further possible factor in explaining divergences in figures for vandalism, wounding and vehicles thefts. Finally, the possibility that more reported crimes are being recorded holds for most offences.

  It is important to assess the statistical significance of these changes. The increases for robbery and mugging were not statistically significant (either between 1997 and 1999, or 1995 and 1999). The only changes in violent incidents between 1997 and 1999 that were statistically significant were those for stranger violence (29% increase) and acquaintance violence (19% decrease).

  

  Click here for a full size version of the chart 

  Closer examination of robbery incidents for 1999 also revealed that there were a few instances where school age respondents had reported multiple incidents of robbery. Although the incidents meet the criterion for robbery, they might be viewed as nearer to bullying incidents. These and other incidents involving school age respondents had a marked impact on estimated changes in violence victimisation rates. Excluding 16-year-olds from the analysis reduced the 14% increase between 1997 and 1999 in robbery to a 2% increase. For the mugging the percentage increase of 4% becomes a decrease of 7%.

  

  Click here for a full size version of the graph