Multilevel study finds no link between minimum wage and crime rates. Article by Dawn Fuller
In general, there are two types of statistical procedures. These include inferential statistics and descriptive statistics. However, the above study is said to have utilized interferential statistics. Ideally, inferential statistics lays its focus on making inferences or predictions regarding a population using the analyses and observations of a sample. The rational for reaching on the conclusion that the above study employed inferential statistics is largely pegged on its mode of analysis: inferential statistical tests. The study employed an econometric time series analysis to analyze its datasets, which is referred to as the autoregressive integrated-moving average (ARIMA).
There are three types of research variables used in the study. These include independent variable, dependent variable and a moderating variable. Also described as an autonomous variable, an independent variable can be defined as variable that can be manipulated to determine the value of other variables in the study. A dependent variable refers to that variable, which are influenced or determined by the states of another variable in the system. Besides, a moderating variable can be described as an interaction that is either quantitative or qualitative, which affect the strength or direction of the relations between independent and dependent variables. In the study, property and crime rate is the dependent variable while the minimum wage rate is the independent variable (Fuller, 2013). Moreover, the consumer price index is the moderating variable in the study.
In the context of statistics, population refers to a set of entities regarding which statistical inferences can be drawn, usually based on a sample. Ideally, the study data is drawn from the entire population consisting of all states in the United States. Sampling is a process employed by researchers to predetermine the number of observations to be taken from a large population. A sampling strategy can either be probability or non-probability in nature. Probability sampling, probability sampling gives every element in the population an equal chance (often greater than zero) to be part of the study sample. This type of sampling strategy makes it possible to produce unbiased estimates of population aggregates, by accurately weighing the sampled units as per their probability of selection. Common used probability sampling strategies include systematic sampling, simple random sampling and stratified sampling. Conversely, non-probability sampling gives the elements of the population with no equal chance of selection. Here, the selection of elements is based on specific assumptions concerning the population of interest. Owing to the definition of the sampling strategy, the above study can be said to have utilized probability sampling (simple random sampling) in selecting of the 18 states (sample) that had raised the minimum wage above the federal mandate at one time between the period in question (Fuller, 2013).
The study employed primary sources of data: economic and crime data from 1977 to 2012. The data on minimum wage and Consumer Price Index was obtained from the United States Department of Labor. Besides, the data on state crime was obtained from Unified Crime Reporting (UCR) program of F.B.I that consists of statistics from law enforcement authorizes within the country (Fuller, 2013). After carefully analyzing the data, the researchers concluded that there is no significant relationship between minimum wage and crime rate among states. Rather, crime is basically an individual-level phenomenon, and the minimum wage does not affect the rate at which individual commit crimes. This conclusion is arguably reasonable based on the study sample, data collected and the method that the researchers used in data analysis.
Reference
Fuller, D. (2013). Multilevel study finds no link between minimum wage and crime rates. Retrieved from http://www.uc.edu/news/NR.aspx?id=18767