Term Project Report: The effect of unemployment of crime Table of Contents Background3-4 Regression Analysis5-6 Conclusion7 References8 Data9 What affects do the unemployment rates have on crime level? 1. Purpose Statement The purpose of this project is to determine how the rate of criminal activity (CRIME) is affected by the rate of unemployment (UNEMP), while also considering the affects of the fluctuation of Consumer Price Index (CPI).
This study uses a time-series analysis with 30 annual observations, from 1969 to 2009. The information was taken from several sources; considering mainly: The Bureau of Labor Statistics and The Bureau of Justice web site. The model is: CRIME = UNEMP+CPI 2. Definition of Variables The dependent variable is defined as the Crime Rate (CRIME). This will take into account the annual rate of both violent and property crimes committed during a measurable time period, of 1970 to 2001.
The independent variables are: • UNEMP (Unemployment Rate) This variable was chosen because when the lack of available opportunity exists to provide income and the means to provide for ones self and/or a family, it puts a negative pressure on an individual to resort to basic survival. We will look at the corresponding annual rate of unemployment over a period of time, from 1970 to 2001. • CPI is Consumer Price Index, which measures changes through time in the rice level of consumer goods and services purchased by households. This variable was chosen because it is a determinant for the level of prices paid by urban consumers for a market basket of consumer goods and services, which turns signifies how out of reach basic goods and services might be for those without resources of income. We will look at the corresponding annual consumer price index over a period of time, from 1970 to 2001.
The relationship between CRIME and the independent variables should be a positive one because when the unemployment rate and the consumer price index decreases, it increase the ability of the households to obtain goods and services, as mentioned before, puts reduces the pressure on the individual to resort to criminal activity. 3. Data Description The data obtained for the purpose of this report was compiled by obtaining a sample of 30 years and obtaining an annual rate for CRIME, which includes both violent and property crimes.
These statistics were obtained from the Bureau of Justice and the online data collection. In addition, the independent variables data obtained include the rate of unemployment and consumer price index for each year selected, which was obtained from the U. S. Bureau of Labor. 4. Regression Analysis The model was calculated using one period = 1 year for the CRIME variable. The results are shown in the table below: Crime = UNEMP(138. 371725) +CPI (2. 291143153) Table 1 Regression Statistics | | | | | | |R Square |0. 383040391 | | | | | |Adjusted R Square |0. 340491453 | | | | | |Standard Error |251. 18294 | | | | | |Observations |32 | | | | | |ANOVA |df |SS |MS |F |Significance F | |Residual |29 |1836041 |63311. 7 | | | |Total |31 |2975950 | | | | |Results |Coefficients |Standard Error |t Stat |P-value | | |Unemployment Rate |138. 3717525 |33. 31553 |4. 15037 |0. 000266 | | |Consumer Price Index |2. 291143153 |1. 06965 |1. 6965 |0. 040725 | | The results of the regression analysis were close to expectations. As projected, there is a positive relationship between CRIME and the independent variables UNEMP and CPI. This indicates that as the rate of UNEMP and CPI (cost of goods) increases, the rate of CRIME will increase. We can also see that there is a stronger statistical significance for the relationship between CRIME and the variable UNEMP, while a weak statistical relationship between CRIME and the variables CPI exist.
In addition, there is a little to no correlation between the independent variables. Table 2 | |Average Crime Rate |Unemployment Rate |Consumer Price Index | |Average Crime Rate |1 | | | |UNEMP |0. 534259935 |1 | | |CPI |0. 128746083 |-0. 14181485 |1 | We can also see through the meaning of t and/or “P-Value” how important the independent variables are to the dependent variable. In this case, the rate of UNEMP shows that the independent variable is very significant. While on the other hand, CPI could be considered a less significant independent variable. The value of Adjusted R2 of 0. 3405 indicates that approximately 34. 1% of the total variation in the dependent variable has been accounted for by the independent variables.
Due to the level of significance of the independent variable, CPI, a second regression calculation was performed in which only CRIME (dependent) and UNEMP (independent) were recorded. The results are as follows: Crime = UNEMP(115. 8516423) Table 3 |Regression Statistics | | | | | | |R Square |0. 285433679 | | | | | |Adjusted R Square |0. 61614801 | | | | | |Standard Error |266. 240114 | | | | | |Observations |32 | | | | | |ANOVA |df |SS |MS |F |Significance F | |Residual |30 |2126513. 49 |70883. 79831 | | | |Total |31 |2975950. 427 | | | | |Results |Coefficients |Standard Error |t Stat |P-value | | |Unemployment Rate |115. 8516423 |33. 46649509 |3. 461720206 |0. 001634512 | |
In this regression where only the variable UNEMP was recorded, the coefficients and the t are positive, which still indicates that as the rat of UNEMP increases, the rate of CRIME will increase respectively. The significance of t tells us that the independent variable does have significance to the dependent variable. However, the value for Adjusted R2 using the single variable decreased to 0. 2616, which shows that only 26. 2% of the total variation in the dependent variable has been accounted for by the independent variable.
This reduction in this value indicates that the single variable might not fit the model as well as with the original two independent variables. 5. Conclusion After evaluating the results of the two regressions, I arrived at the following conclusions: 1. The results on the first regression should not be improved. Despite the fact that both outputs show a positive relationship between the independent and dependent variable, the positive relationship of the variables as well as the Adjusted R2 did not improve at all in the second regression. 2.
In fact the Adjusted R2 actually decreases significantly in the second regression, indicating that the data does not fit the function as well. 3. A positive correlation between the rate of CRIME and UNEMP and CPI existed according to the results on the first regression even though CPI seemed to have less significance over the dependent variable. So, it might not affect the rate of the CRIME as much as the variable UNEMP does. 4. While the low R-Square value indicates additional variables could be added to strengthen the model we can assert that the rate of CRIME is influenced by the rate of Unemployment and the Price of Goods. . Works Cited Page 1. Economic Report of the President 2010 pages 371-410. Retrieved on September 15, 2010 from: http://www. whitehouse. gov/sites/default/files/microsites/economic-report-president. pdf 2. Historical US Unemployment and Consumer Price Index Data Table Retrieved on September 9, 2010 from: http://www. bls. gov/data/ 3. Historical US Crime Rate Data Table Retrieved on September 9, 2010 from: http://bjs. ojp. usdoj. gov/dataonline/ 4. Raphael, S. & Winter-Ebmer R. . “Identifying the Effect of Unemployment on Crime. ” Journal of Law and Economics 7. Data (see attached worksheet) [pic]