TABLE 2
FROM:
Adding the Temporal and Spatial Aspects of Routine Activities: A Further Test of Routine Activity Theory
Elizabeth R Groff
BACK TO ARTICLETable 2. Parameters in the model
| Variable | Rationale |
|---|---|
| Society level | |
| Number of agents = 1,000 | Represents a balance between ensuring there are enough agents so that interactions can occur and the computational overhead from using more agents. |
| Number of police = 200 | Chosen to ensure that cops would be present at some of the convergences that occur across the 16,035 places in Seattle. The actual ratio is 1 to 472. Subsequent tests with two police did not change the main conclusions of the research. |
| Unemployment rate = 6% | The unemployment rate of 6% is based on the 2002 unemployment rate for Seattle (Bureau of Labor Statistics, 2003). |
| Rate of criminal propensity = 20% | Given that 20% of the population has committed a crime, 20% of civilians are assigned criminal propensity using a uniform distribution (Visher and Roth, 1986). This value is a starting point. Subsequent tests using 5 and 90% criminal propensity did not alter the main conclusions of the research. |
| Time to reoffend = 60 | Parameter value chosen as a starting point since the author could find no empirical data on which to base time to reoffend |
| Random number seed = 100 (seed also tested at 200, 300, 400 and 500) | An explicit random number seed based on the Mersenne Twister (MT) algorithm is used as the basis for all random number distributions used in the model. MT is currently considered to be the most robust in the industry (Ropella et al, 2002). |
| Agent level | |
| Societal time spent away from home = 30% (40, 50, 60, 70%) | Assigned based on a normal distribution with a mean of 432 minutes (for the 30% condition) and a standard deviation of 10% of the mean (s.d.=43). |
| Initial wealth = 50 | Initial wealth is assigned with a mean of 50 and a standard deviation of 20 units. |
| Amount of wealth received each payday = 5 | No empirical evidence available. |
| Amount of wealth exchanged during robbery = 1 | No empirical evidence available. |
| Situation level | |
| Guardianship perception = U(-2,2) | The guardianship perception value can add or subtract zero, one or two guardians from the actual number present. This represents the stochastic element in the offender's perception of the willingness of a guardian to intervene. |
| Suitable target perception = U(-1,1) | The value of a suitable target can increase or decrease the suitability or leave it unchanged. This enables the offender to sometimes decide a target is not suitable even when they have more wealth. |
