Trends in Recorded Crime

• NCJ Number: 73790
• Journal: Delikt en delinkwnet Volume: 9 Issue: 5 Dated: (1979) Pages: 303-310
• Author(s): H Angenent
• Date Published: 1979
• Length: 8 pages
• Publication Format: Article
• Publication Type: Statistics
• Country: Netherlands

Annotation

Two methods for calculating directions of crime rates are described.

Abstract

Trend designates a general impression of the direction in which crime rates are moving, in particular whether the rates are increasing or decreasing as represented by the most appropriate straight line. In a graph with years recorded on the horizontal x-axis and the crime rate on the vertical y-axis, the line joining the plotted yearly rates will be a zig-zag. The trend of the zigzag line is plotted by drawing a horizontal line through the average x average y point and determining the distance (d) from the points of the zigzag line to this line. A measure of variation is the sum of the squares of these distances. The most appropriate straight line is the line for which the sum of the squares of the distances is minimal and to which the average points are significantly closer than to the horizontal line. An F-test establishes whether the trend is significant. However, the course of crime is measured with the most appropriate curved rather than the most appropriate straight line. The most appropriate straight line is a first degree function with the formula y equals ax plus b. The most appropriate second degree function is a curve with the formula y equals ax squared plus bx pus c, and the formula for the third degree function is y equals ax squared plus bx squared plus cx plus d. The highest function for n-years is a (n-1) degree function. Usually it is best to stop at the fourth degree function. The same methods are used to test the most appropriate function as to determine the most appropriate line. The course of crime is represented by the lowest function that fits the direction significantly better than the horizontal straight line. Sometimes the trend and the course cannot be determined in this manner and must be represented in another fashion, as several brief examples illustrate. Figures and a 3-item bibliography are provided.