The Effect of Unemployment on Rape - Research Paper Example

A Correlation Analysis of the Effect of Unemployment on Rape Research question/hypotheses Research Question: Is there a relationship between rape andunemployment? Null Hypothesis: There is no statistically significant relationship between the rates of rape and unemployment. Alternative Hypothesis: There is a statistically significant relationship between the rates of rape and unemployment Research Design and Sample The aim of this study is to test for a significant relationship between the rate of rape and rate of unemployment. To do this rape and unemployment data from each of the 50 states will be analyzed for the year 2010. Data Collection The independent variable for this study was the unemployment rate. This is a percentage value calculated from the division of the number of unemployed people by the labor force. Someone is considered to be unemployed if they are above 16 years of age who had no employment during the period but was available and desired to work. In contrast an individual is employed if they are over 16 and worked for at least one hour during the reference period. In general the reference period is one week. The labor force is the total number of people considered to be unemployed or employed, For this study statistics on the unemployment and employment levels for each of the 50 states were analyzed for the 2010 period. Information was gathered from the Business First database (available at: http://www.bizjournals.com/buffalo/datacenter/state-employment-2010.html). The database gave the average number of people employed and unemployed for the year 2010; this was derived from information collected by the Bureau of Labor Statistics how were the data collected? be more specific here and as a consequence represented the mean value from 12 months of data. As a consequence, the unemployment rate could be calculated by adding the employment and unemployment levels together (to provide the labor force value) and then dividing the unemployment level by the labor force to give a percentage of the labor force which was unemployed. This value is the unemployment rate. The unemployment rate was determined independently for each of the 50 states. The dependent variable for this study was forcible rape per 100,000 population. For this, the FBI crime statistics from 2010 were used (available at: http://www.fbi.gov/about-us/cjis/ucr/crime-in-the-u.s/2010/crime-in-the-u.s.-2010/tables/10tbl05.xls). These statistics considered many different types of crime including forcible rape in each of the states and included the number of reported incidences of rape as well as the estimated total of rape cases. The estimated total number of rapes was used for this study. Aggravated rape is defined as is the forcible sexual penetration of a woman as well as assault with this intent or attempted but not successful rape. Each incident of either rape or attempted rape was recorded as an offense within the database. Statutory rape was not included in the database and as a consequence was not part of this study. Analysis As there is no experimental manipulation of either variable, cause and effect cannot be directly tested, consequently, the analysis needs to test for correlation. In order to do this, a statistical test for correlation needs to be undertaken. To test whether there is a significant relationship between rape and unemployment rate a Pearson Product-Moment Correlation will be conducted. This test looks for a linear relationship between two variables and can describe a positive relationship, negative relationship or absence of relationship. The Pearson correlation coefficient is known as r and is a measure of how closely the data points are to the line of best fit that is described by the Pearson Product-Moment Correlation. A value of below zero indicates a negative correlation between the two variables (in this case rape and unemployment), a value above zero indicates a positive relationship, and a value close to zero indicates an absence of relationship (Siegel, 1957). This form of analysis is used because it is a correlation test for ratio or interval variables (Jakobsson, 2004). This is relevant to the data as the information for rape from the FBI database is an interval variable, while the data for unemployment rate is a ratio. As a consequence, other forms of correlation analysis such as Spearman’s Rank Order Correlation are not relevant. For both variables statistical information was gathered for each of the 50 states. In the analysis, each of these states will form a data point. Feasibility There is a significant amount of information on both rape and unemployment rates for all 50 states for the year 2010. This should provide sufficient information for the statistical analysis and result in a reliable outcome. The information was easy to find. Significance It is expected that the statistical analysis will show no relationship between the rate of unemployment and the amount of rape. This prediction is based off a 2001 study which looked at the correlation between rates of victimization for the major types of crime and the unemployment rate among civilians. When specifically examining rape the study found that there was a weakly positive relationship between the amount of rape and the level of unemployment. This effect was not statistically significant. The estimates of rape used for this study overestimated the level of rape, and as a consequence the authors concluded that the slightly positive trend they observed was a consequence of that, and as such it was likely there was no actual affect of unemployment (Raphael & Winter-Ebmer, 2001). However, this analysis uses a different set of information for the level of rape, and in addition the study year is later than that of the study and as a consequence it is likely that the information will be of greater accuracy. Consequently, this analysis may find a different conclusion than the study. The results of this proposed study may have strong implications for understanding a factor contributing to rape and potentially reducing rape. If it is found that there is a significantly positive relationship between the level of unemployment and the amount of rape, then this indicates that strategies that are used to increase the level of employment may also be of relevance to decreasing levels of rape. In addition, this suggests that the ideas proposed in the paper by Raphael and Winter-Ebmer (2001) may be correct. These ideas were that, the amount of rape that occurs within society can be in part linked to the amount of time that perpetrators have available to them. In addition, they propose that the being in employed decreases the desire to commit crime. Strengths and Limitations One of the strengths of this analysis is that there is sufficient data to ensure a thorough and complete examination of the relationship between the two variables. In addition, the data that are used in this study are recent. This is a strength as it means that the conclusions that are drawn from the study are relevant to current society and reflect trends that are occurring now, not in the past. Also, the data from both sources is given as a mean for the 2010 year, which makes analysis simple and comparable between the two variables. Finally, the test that is being used is relevant for this type of data and consequently will give a reliable result. One limitation of the analysis is the statistical method that is employed. Firstly, the statistical method that is being used tests for a correlation between two sets of variables. As a consequence, both variables are treated equally, and it cannot be determined what variable comes earlier in time (cause and effect). The result of this is that the results of the analysis only show whether there is a relationship between the factors and not whether one is driving the other. This means that potentially, although it is difficult to determine a method in which this could occur, rape could be the driving factor as opposed to unemployment. A more likely alternative is that both variables are being driven by an external factor. If this was the case, the test would show a significant relationship between the two variables, and it would appear identical to if rate of unemployment was influencing rape. For example, the economic environment has been harsh in recent years, and as a consequence living has become harder for many people. The consequence of this has been that many companies have had to cut down on the number of staff that they hire, increasing the unemployment rate. At the same time, the bad economic environment might mean that people are more desperate for pleasure, and more stressed and angry that they cannot get help. This could increase the amount of rape. Consequently, even if a correlation analysis would show that there is a significant relationship between rape and unemployment the relationship is an artifact of both being modified by an external factor, and thus is not an indication of unemployment affecting rape. A second limitation is that the analysis considers only data from a single year. As a consequence both the data and the analysis provide a snapshot in time. The study may find a relationship (or absence of a relationship) between the variables, but it is difficult to extrapolate from this information to general trends, as there is no way of determining whether the particular year of study was abnormal. For example, there may have been a worldwide event in 2010 that affected both unemployment rate and the amount of rape that occurred which is unusual, and as a consequence the same trend is not applicable to other years. The results cannot be generalized to other times and places. Additionally, data points for the correlation analysis are yearly averages or each state, with one measure of amount of rape and one measure of unemployment per state for the 2010 period. The statistical purpose of this is to provide enough replication to produce an accurate analysis; however, this relies on the relationship between rape and unemployment being similar across the different states which may not be the case. States differ significantly in environment, in expectations, in many laws as well as other factors. As a consequence, it is not reasonable to assume that the relationship between the two variables will be the same in each state (Stevens, 2002). Thus, the determined trend may not represent a true picture of the relationship between rape and unemployment. The type of data used for this study limits the relevance of any conclusions that are drawn. One issue with using data based on the individuals reporting rape is that many people do not report being raped, and the rate of reporting can be influenced by many different factors, such as the socioeconomic position of the victim (Clay-Warner & McMahon-Howard, 2009). In addition, an instance of rape for the data that was analyzed also included attempted rape, but did not include statutory rape. As a consequence, the results cannot truly be generalized to all forms of rape. Finally, this form of analysis makes assumptions of the data, which must be fulfilled in order for the results of the analysis to be relevant. These are that both variables are in the form of ratio or interval measurements, that there is normal distribution, that outliers are removed or reduced, that the data is homoscedastic (the variation that is within the data collected is the same) and a linear relationship is present between the two variables. If the data does not correspond fully to all of these assumptions, then the outcome of the test is not accurate (Havlicek & Peterson, 1977). As there are a limited number of different tests, alternatives such as transforming data must be considered if the data does not meet the assumptions (Siegel, 1957). References Clay-Warner, J., & McMahon-Howard, J. (2009). Rape Reporting: Classic Rape and the Behavior of Law. Violence and Victims, 24(6), 723-743. Havlicek, L. L., & Peterson, N. L. (1977). Effect of the violation of assumptions upon significance levels of the Pearson r. Psychological Bulletin, 84(2), 373. Jakobsson, U. (2004). Statistical presentation and analysis of ordinal data in nursing research. Scandinavian journal of caring sciences, 18(4), 437-440. Raphael, S., & Winter-Ebmer, R. (2001). Identifying the effect of unemployment on crime. JL & Econ., 44, 259. Siegel, S. (1957). Nonparametric statistics. The American Statistician, 11(3), 13-19. Stevens, J. (2002). Applied multivariate statistics for the social sciences: Lawrence Erlbaum, Read More