Next | Previous | Contents | Taxi-L Home
|
|
| Canadian Taxi Driver Homicides, 1917-2007 |
|
Next | Previous | Contents | Taxi-L Home |
|
|
2E. Day of Year
Contents
Category A Homicides by Day of Year (Chart)
As noted under Month / Day of month, this page looks for a time-related pattern in the occurrence of homicides. Once again the data reflect spikes in the number of homicide occurrences at certain times of the year. But the data also show that homicides can occur at any time of the year, so it is hard to come to any practical conclusions so far as driver safety is concerned. Chart 2E-1 represents the day of occurrence for 147 Category A homicides for which we have precise date information:
It appears that 54 of the 147 homicides (37 percent) occurred during the 86 days between day 260 (September 17) and day 345 (December 11). The "Day of Year" data seem to give a more precise indication of the seasonality of past homicides than do the data for cant2c.htm">Homicides by Month, but of course past history does not necessarily reflect current or future trends. A word of caution: although the significant factor is the number of homicides in a cluster, the chart tends to emphasize clusters that span several days. As a result the 11-homicide spike occurring over the short period between Days 95 and 104 appears much less dramatic than the 18-homicide clusters. The procedure for creating Chart 2E-1 is described below under Homicide Clusters.
Category A Homicides by Day of Year (Table)
Table 2E-2 provides the numbers represented graphically in Chart 2E-1 (right click on the link to see the table in a new window):
Each day is numbered consecutively from 1 (January 1) to 365 (December 31). The leap year date of January 29 is combined with March 1 as Day 60 (no homicides have actually occurred on January 29). One homicide (Roland Faucher) was excluded because we do not yet have precise date information.
Homicide Clusters
Chart 2E-3 illustrates the process of identifying clusters of homicides in order to determine whether attacks have occurred more frequently at certain times of the year.
The chart represents a "close up" of the range from Day 44 (February 13) to Day 65 (March 6). The red blocks represent individual homicides which occurred on Days 46, 49, 52, 54, 56, 59 and 63. Two homicides occurred on Day 59. If we group homicides that have no more than a one-day interval between them we identify the cluster represented by the blue rectangle. The blue rectangle is a graph of three homicides occurring over five days (Days 52 to 56). If we allow an interval of two days between homicides, we identify the cluster represented by the yellow rectangle. The yellow rectangle is a graph of seven homicides occurring over 14 days (Days 46 to 59). If we allow an interval of three days between homicides, we identify the cluster represented by the green rectangle. The green rectangle is a graph of eight homicides occurring over 18 days (Days 46 to 63). Chart 2E-1, representing 147 homicides over 365 days, shows clusters spanning intervals of no more than three days (similar to the green rectangle in Chart 2E- 3).
|
