"By fixing the control variables at their means and adjusting only the number of slaughterhouse employees in a county it is possible to see how different levels of slaughterhouse employment would affect the scales (see Table 5). An average-sized slaughterhouse, which employs 175 people, would be expected to increase the arrest scale by 2.24 arrests and the report scale by 4.69 reports. Particularly telling is the fact that the expected arrest and report values in counties with 7,500 slaughterhouse employees are more than double the values where there are no slaughterhouse employees. These results demonstrate that the effect of slaughterhouse employment on these scales cannot be explained away by the control variables and that the comparison industries do not have similar significant effects. Also, because the analyses employ fixed effects they also therefore control for time-invariant variables in these counties that might affect the crime rates, such as geographic location. These findings, however, cannot provide insight into how slaughterhouses, the comparison industries, and the control variables affect individual crime variables. To provide this insight, we used negative binomial regression."The arrest scale goes from 69.32 to 71.56, the report scale from 115.40 to 120.09.
"For some analyses, crime rate variables were created and factor analyzed to create two scales (arrest rate and report rate scales). Using the scales as dependent variables mitigates the variables mitigates the violations of OLS regression assumptions by creating a more normal distribution of scores than obtained with the counts or rates for particular crimes. To create the scales the counts were first converted into rates. Then principal components analysis was used to determine the factor structure, followed by iterative principal factors to obtain the factor loadings. The resulting Arrest Rate Scale is made up of the following variables: rape, robbery, burglary, other assaults, forgery, possessing stolen property, vandalism, offences against the family, and disorderly conduct.8 The same process was followed to create the Report Rate Scale.9 The Report Rate Scale is made up of the following variables: reports of rape, robbery, assault, burglary, motor vehicle theft, and arson.""8. The factor loadings are all above the commonly accepted minimum values of 0.3 to 0.4 and the Chronbach’s alpha for the scale is .6728.
« Older Why, in this day and age when you can get them for... | The Intor.net - A userless, in... Newer »
This thread has been archived and is closed to new comments
posted by empath at 10:17 AM on May 15, 2010