Article Critique

"STRIKING OUT" AS CRIME REDUCTION POLICY: THE IMPACT OF "THREE STRIKES" LAWS ON CRIME RATES IN U.S. CITIES

Identify the research design used

The research design used in this paper is Multiple Time Series (MTS). In order to understand Multiple Time Series, first we have to comprehend Time Series research design. Time Series is a research design in which same variables are measured through a varying period, often with a purpose of determining social trends. In this research design, measurements are taken continuously which differentiates it from 'one shot' designs. Official data can be used in this research design; as crime rates can be plotted for different times of the same area. The Multiple Time Series Design includes an equivalent control group, which is absent in Time Series. One of the main reasons for using MTS here is that a more thorough experimental control was not practical (Campbell & Stanley, 1963, pp. 55-57).

Specify the type of data used to conduct the analysis

The data used to conduct the analysis is Uniform Crime Reports (UCR). UCR is the official data on crime records in USA, published by FBI. The crime statistics in the UCR is published annually. The law enforcing agencies collect the data, which is then provided to the FBI. This study used UCR data from 188 cities, spanning the period of 1980 to 2000. City-level data was utilized which spanned a period of two decades. A total of 188 cities were included for data acquisition, with 110 three strikes states and 78 control cities, where the law was not enforced. This large sample size gives more statistical power and the capability to identify more subtle effects of the law on crime rates (Wooldridge, 2000, p. 409).

Identify the sampling procedure employed

In this study, a very large sample of 188 cities was used. The sample employed 'three strikes' as well as control cities. The advantage of using control cities was that if the law was successful and there was a greater drop in crime rates in the states where the law was passed, then the difference could be spotted and identified. This sample had many advantages. It had the flexibility of entering proxy variables. The large sample size enabled them to use various controls for factors that might be connected with other variables and hence result in false associations (Wooldridge, 2000, p. 434). The reason for using the city as the unit is that ...