Crimes in Kenosha: What are the Trends? By Alexandria Caputo An Undergraduate Thesis Submitted in Partial Fulfillment for the Requirements of Bachelor of Arts In Geography and Earth Science Carthage College Kenosha, WI December 2014 1|Page Crime in Kenosha: What are the Trends? Alexandria Caputo December 2014 Abstract Crime has been studied for as long as crimes have been committed. There is a solid base of information available that portrays common spatial patterns and trends when analyzing crime. Conducting hotspot analyses of five different crime type, including burglary/theft/robbery, domestic violence, assault/battery, disorderly conduct and drug/alcohol crimes. Often time police have limited time and resources in order to help keep citizens safe, so analyzing where hotspots are located throughout the Kenosha boundaries can help police focus their efforts on specific areas in which clustering occurs. Based off these hotspots, overlaying crime layers with different socioeconomic variables can show different trends as well. The Kenosha Police Department can use this data to gain a better understanding of crime patterns in their jurisdiction and with that they can develop methods of prevention. Acknowledgements I would like to thank my thesis advisor Dr. Wenjie Sun for helping guide me in writing my senior thesis, always giving me feedback on my drafts and maps, and providing me with the necessary resources to complete my thesis well. I would also like to thank Dr. Kurt Piepenburg for assisting wherever I needed him and making sure deadlines were met. I would like to thank Dr. Matt Zorn and Dr. Joy Mast for helping me grow in my education throughout my first few years at Carthage. Without the knowledge I obtained from being in your class, I would never have been given opportunities that I have. 2|Page Table of Contents List of Tables….................................................................................................................4 List of Figures…………………………………………………………………………….…..…5 Introduction…………………………………………………………………………………...…6 Problem Statement…………………………………………………………………………......6 Hypotheses…………………………………………………………………………………..7 - 8 Hotspots ...............................................................................................................7 Demographic Factors……………………………………………………………….7 - 8 Literature Review……………………………………………………………………….….8 - 12 Crime Pockets……………………………………………………………………..……8 Seasonal Trends in Crime Rates……………………………………………...…9 - 10 What Causes Differing Crime Rates………………………………………….. 10 - 12 Methods …………………………………………………………………………….…....12 - 14 Site Selection………………………………………………………………………….12 Data Acquisition………………………………………………………………………13 Converting into Usable Layer File……………………………………………….….13 GIS: Mapping………………………………………………………………………….13 GIS: Collect Events Tool……………………………………………………………..13 GIS: Average Nearest Neighbor ……………………………………………….13 - 14 GIS: Getis-Ord GI Test…………………………………………………………….….14 Overlay Demographic Data with Hotspot Layers……………………………….….14 Results…………………………………………………………………………………....14 - 42 Distribution of Crimes……………………………………………………….…..15 - 21 Frequency of Crimes……………………………………………………….…...22 - 26 Average Nearest Neighbor ………………………………………………………….27 3|Page Hotspot Analysis…………………………………………………………………..28-35 Demographic Maps………………………………………………………..…….36 - 42 Discussion and Conclusion ………………………………………………………………….43 Future Research……………………………………………………………….….43- 45 References………………………………………………………………………………….….45 List of Tables Table 1: Average Nearest Neighbor Results 4|Page List of Figures Figure 1: Distribution of Crime Figure 2: Distribution of Burglary/Theft/Robbery Crimes Figure 3: Distribution of Domestic Violence Crimes Figure 4: Distribution of Drug and Alcohol Crimes Figure 5: Distribution of Assault and Battery Crimes Figure 6: Distribution of Disorderly Conduct Crimes Figure 7: Frequency of Burglary/Theft/Robbery per Location Figure 8: Frequency of Domestic Violence Crime per Location Figure 9: Frequency of Drug and Alcohol Crimes per Location Figure 10: Frequency of Assault and Battery Crimes per Location Figure 11: Frequency of Disorderly Conduct Crimes per Location Figure 12: Hotspot Analysis of Burglary/Theft/Robbery Crimes Figure 13: A Close-up Hotspot Analysis of Burglary/Theft/Robbery Crimes Figure 14: Hotspot Analysis of Domestic Violence Crimes Figure 15: Hotspot Analysis of Drug and Alcohol Crimes Figure 16: A Close-up Hotspot Analysis of Drug and Alcohol Crimes Figure 17: Hotspot Analysis of Assault and Battery Crimes Figure 18: Hotspot Analysis of Disorderly Conduct Crimes Figure 19: A Close-up Hotspot Analysis of Disorderly Conduct Crimes Figure 20: Disorderly Conduct Hotspots VS Average Household Income Figure 21: Drug and Alcohol Hotspots VS Average Household Income Figure 22: Disorderly Conduct Hotspots VS Diversity Index Figure 23: Drug and Alcohol Hotspots VS Diversity Index Figure 24: Disorderly Conduct Hotspot VS Percent of Male Population Over 18 Figure 25: Drug and Alcohol Hotspot VS Percent of Male Population Over 18 5|Page Crimes in Kenosha: An Analysis of Spatial Patterns and Trends Alexandria Caputo Introduction Crime has been studied for as long as crimes have been committed. There is a solid base of information available that portrays common trends found when analyzing crime. Often time police have very limited time and resources in order to help keep citizens safe. While it is important to understand crime trends on a national scale, not all patterns are the same. Police departments should understand the crime trends for their jurisdiction, in order to best predict and respond to crimes. What might work for one department might not work for another one; therefore police need to be well-educated on trends that can occur in their area. The crime data available for the study area was for a five month time span during the warmer months. It is important to note that crimes might follow a specific temporal trend, therefore the study conducted focused on crimes that do not tend to follow a seasonal trend, as data from all seasons was not available. Problem Statement Crimes often do not distribute randomly or uniformly in space, rather there are pockets of more dangerous areas. It is important to find and target limited police resources at the crime hotspots in order to more effectively and efficiently fight crime. Demographic and socioeconomic factors may have an impact on the spatial distribution of crimes. 6|Page Hypotheses Analyzing crime hotspots i.e. clustering in spatial data, can portray where crimes are located throughout the Kenosha boundaries. Crime Hotspots 1. · · All crimes combined H01 – There is no hotspot in the spatial distribution of all types of crime combined. HA1 – There are hotspots in the spatial distribution of crimes combined. 2. · · Burglary/Theft/Robbery HO2 – There is no hotspot in the spatial distribution of burglary/theft/robbery crime. HA2– There are hotspots in the spatial distribution of burglary/theft/robbery crime. 3. · · Domestic Violence HO3 – There is no hotspot in the spatial distribution of domestic violence crime. HA3 – There are hotspots in the spatial distribution of domestic violence crime. 4. · · Drug and Alcohol HO4 – There is no hotspot in the spatial distribution of drug and alcohol crime. HA4 – There are hotspots in the spatial distribution of drug and alcohol crime. 5. · · Battery HO5 – There is no hotspot in the spatial distribution battery crime. HA5 – There are hotspots in the spatial distribution of battery crime. 6. · · Disorderly Conduct HO6 – There is no hotspot in the spatial distribution of disorderly conduct crime. HA6 – There are hotspots in the spatial distribution of disorderly conduct crime. Demographic Factors 7. Income · HO7 – Crime rate is not lower in higher income areas. · HA7 – Crime rate is lower in higher income areas. 8. Diversity · HO8 – Crime rate is not lower in lower diversity areas. · HA8– Crime rate is lower in lower diversity areas. 7|Page 9. Percent of Males over 18 · HO9 – Crime rate is not higher in areas where there is a higher percent of males who are above 18 years old. HA9 – Crime rate is higher in areas where there is a higher percent of males who are above 18 years old. Literature Review Crime Pockets When studying crime rate trends in a particular area, there are bound to be areas where there are crime pockets. Crime pockets are hotspots where there are more crimes in a section of a neighborhood, compared to other parts of the neighborhood. Peter K. B. St. Jean studied the geography of crime in Chicago, focusing mainly on the “Bronzeville” neighborhood. Through his research he found that there were certain “hotspots” in the study area. These “hotspots” are locations in which a significant amount of dangerous activities cluster, in comparison to the rest of the area studied. In these “hotspots”, St. Jean found that there were higher poverty rates and high residential turnover, compared to other locations. There is a lack of neighborhood cohesion among the neighbors, because of the high residential turnover which causes a lack of trust and expectations among them. (St. Jean 2007). Hotspot mapping is a really popular analytical technique which is used to identify where the police should target their crime reduction resources. Hotspot mapping relies on previous data to identify high concentrations of crime. Police use the hotspots to predict where crime is most likely to occur in the months ahead. Hotspot mapping accuracy depends on the type of crime. Hotspot mapping was better in predicting future 8|Page street crimes than any other crime type however; it was still able to predict the future of other crimes as well (Chainey 2008). Seasonal Trends in Crime Rates When looking at crime trends there is already a huge base of knowledge available. When doing a study of crimes where there is limited data available, making sure that you study select crimes that are not affected by seasonal changes is very important. Crimes that are affected by the changing of the seasons tend to see a cyclic pattern with spike or drop in rates throughout the year. There are many things that can explain this seasonal change in crimes. There are two main theories that describe seasonal changes in crimes: Temperature/aggression theory and routine activities theory. Both theories were first proposed by Quetelet. The temperature/aggression theory states that “uncomfortably hot temperatures increase the frustration that individual’s feel, which leads to aggressive behavior.” The routine activities theory states that when there are pleasant temperatures outside, more individuals spend less time in their house and more time outdoors. They have more time to spend on crimes as a result of this (Quetelet 1969). Since Quetelet did his research in France, more recent studies are comparing the patterns he saw in France to what they are finding in the United States. A group of four professors from University of North Carolina at Chapel Hill and North Carolina State University dedicated their efforts to testing Quetelet’s two theories. They found that each of the two theories suggests a positive relationship between seasonal temperature and changes in crime rates. They found that the temperature/aggression theory does 9|Page show that hotter temperatures correlate to greater aggression and violent crimes. However it does not affect property crime rates because property crime tends to involve calculating behavior and not necessarily aggression. They have found that in the routine activities theory that people are more likely to be outside when the weather is nicer. With more time spent outside the home, there is more time to commit crimes. They found that this theory however, does not mean an increase in aggressive crimes (Hipp et al 2004). Two professors at Wayne State University in Detroit Michigan through their research have found that crimes such as burglary, larceny vehicle theft, and arson show no changes in rates with the changing seasons (Lim et al 2009). What Causes Differing Crime Rates There are many different factors that affect crime rates throughout the world. There have been many studies that examine crimes at a national level. A main factor that has been found to affect crime rates is inequality. Rosemary Gartner studied the relationship between crime and inequality on a national scale. Her studies have found that countries with a high degree of economic inequality have a higher level of violence. She proposed that the nation’s homicide rate is shaped by four different contexts: material, integrative, demographic, and cultural. Homicide rate is shaped by material functions. There is a form of economic stress that results from an unequal distribution of resources. When there is a family that has absolute deprivation from resources, there is an increased risk of homicides when being compared to their counterparts. If there is economic inequality, she found that this results in aggression and hostility. She also found that people tend to kill those who they feel in competition with. The government needs to get more involved in certain places: “Where government efforts provide a 10 | P a g e minimum living standard are more limited, and where income inequality is greater, homicide rates will be greater for both females and males, adults and children.” The next context that Gartner proposed was the rates differed based on social integration. Weakened social integration lessens social control in the individual, which could increase most types of homicide. People that integrate themselves into society really well, tend to be able to control themselves in different environments and therefore less likely to commit any form of homicide. Demographic factors also affect changes in homicide levels. If a population is younger, then their activities are dispersed and people are involved in so many things. Because of this they are less effective in controlling themselves socially, and the opportunity for homicide rises. Culture is another thing that affects rates of crime. If there are a lot of varying cultures that live in close proximity to each other, there is more likely to be violence as a way to try to solve it (Gartner 1990, 1992). There was another study that focused on the links between economic inequality, family disruption, and urban African- American violence in more than 150 cities in the United States. This study shows us that as economic inequality increases throughout an area so do arrests of African-American youths for violent crimes. If an area has a higher income inequality, this tends to increase the number of African-American single-parent households, which is related to the increase of youth violence. Single parents have more stress and fewer resources which causes them to have a more difficult time monitoring their children (Shihadeh et al 1994). We live in a world where people are focused on having the newest and “next big thing” on the market. Unfortunately this causes a lot of trouble amongst people who do 11 | P a g e not have the appropriate monetary means to materialistic goods. There is this attitude that you have to pursue the acquisition of goods by any means necessary, even if how you achieve them is illegal (Kramer 2009). For the economically disadvantaged people “this anomic orientation leads not simply to high levels of crime in general but to especially violent forms of economic crime, for which the United States is known throughout the world, such as mugging, car-jacking, and home invasion” (Messner 1997). Another factor that affects crime rates is disruptive families. Quetelet found that if a family has at least one member who actively engages in disruptive behavior, the family is less likely to feel like they belong in a neighborhood. These families tend to leave each time a member becomes disruptive in a neighborhood, to areas in which rent is lower. This tends to cause concentrations of disruptive families because they all find cheap rental properties, therefore a crime pocket occurs. The crimes that tend to occur when disruptive families live in one area are domestic violence and acquaintance battery (Quetelet 1969). Methods Site Selection For this study, Kenosha, Wisconsin was chosen. Kenosha is located just north of the Wisconsin-Illinois border between Milwaukee, Wisconsin and Chicago, Illinois. As of 12 | P a g e 2013, Kenosha has a population of 167,757 people. Eighty-Seven percent of the population are of the white race (US Census Bureau, 2013?). Data Acquisition Crime data was acquired from the Kenosha Police Department for the period of May 1, 2014 until September 30, 2014. The information included type of crime, date and time committed, and location. The data obtained came in the form of a spreadsheet with over 1600 crime records available. The Excel spreadsheet was then formatted in order to be used in ArcGIS software. The types of crimes analyzed were assault/battery, domestic violence, burglary/theft/robbery, disorderly conduct, and drug/alcohol crimes. GIS: Mapping Mapping is commonly used to visualize spatially-tagged data and portray spatial relationships. A simple data table of all the crimes committed would not be enough to capture the spatial pattern of crimes that occurred in Kenosha throughout the five month time period. Mapping out the crimes would provide a visual representation that would be easier to understand than a data table. GIS: Collect Events Tool Some locations/addresses have multiple crimes happened during the 5 month time period. This tool graphed out where these locations were using graduated symbols. The bigger the marker on a point was, the more crimes that were committed at that location. There were as few as 1 crime at an address and as many as 14 crimes so this 13 | P a g e graduated symbol map portrayed where crimes happened more often. The five types of crimes were all analyzed separately and then together. GIS: Average Nearest Neighbor The Average Nearest Neighbor is a spatial statistic tool in ArcGIS. This tool calculates a nearest neighbor index based on the average distance from each feature to its nearest neighboring feature together with a z-score, which indicates the level of statistical significance. The nearest neighbor index is then compared to that derived from a hypothetical random distribution with the same number of features (crimes) and the same study area, to infer whether the global pattern is clustered, dispersed, or random. GIS: Getis- Ord Gi* (Hotspot Analysis) The Getis-Ord Gi* test identifies local clusters of high or low feature attribute values. This test looks at individual locations and their attribute values, which in this case is the number of crimes and then compares the value to those of neighboring addresses and eventually identifies statistically significant hot spots (clusters of high values) and cold spots (clusters of low values) using the Getis-Ord Gi* statistic. The five types of crimes were all analyzed separately so the trends can be seen for each different crime type, and then together to analyze general trends in crime. Overlay Demographic Data with Crime Hotspot Layers Mapping out demographic data, including income, diversity, and percent of males over 18, with the crime hotspot layer on top will show possible spatial correlations between demographics and crime rates. 14 | P a g e Results After processing the data, visually portraying the data, and interpreting what it all meant, it showed that there were spatial trends in certain crime types. My hypotheses were accurate in the Burglary/Theft/Robbery, Drug and Alcohol, and Disorderly Conduct crime types, in that they all portrayed hotspots of dangerous activity. My data showed that my hypothesis was incorrect because assault/battery and domestic violence did not portray any hotspots. My data supported my hypothesis that crime is not lower in higher income areas, but does not support my hypotheses of crime being lower in lower diverse areas, or crime is not higher in areas which have a higher population of males over 18. Distribution of Crimes The following maps (Figure 1-6) show the distribution of crimes throughout Kenosha. 15 | P a g e Figure 1: Distribution of All Crime Around 1600 crimes were committed in Kenosha from May 1, 2014 through September 30, 2014. 16 | P a g e Figure 2: Distribution of Burglary/Theft/Robbery Crimes The above map shows where 531 burglary/theft/robbery crimes were committed throughout the five month time span. There are places where crimes are concentrated, while there are places where there is one crime secluded from the others. 17 | P a g e Figure 3: Distribution of Domestic Violence Crimes The above map portrays the distribution of 111 domestic violence crimes committed in the Kenosha area. 18 | P a g e Figure 4: Distribution of Drug and Alcohol Crimes The above map is a visual representation of the distribution of 326 drug and alcohol crimes. 19 | P a g e Figure 5: Distribution of Assault and Battery Crimes The above map portrays the distribution of 104 assault and battery crimes. 20 | P a g e Figure 6: Distribution of Disorderly Conduct Crimes The figure above shows the distribution of 474 disorderly conduct crimes committed in Kenosha in the five month span. 21 | P a g e Frequency of Crimes The frequency of crimes per location portrays that some addresses have multiple crimes that happen during the five month time frame. The following figures (Figure 7- 11) portray the frequency of crimes occurring at a particular address. 22 | P a g e Figure 7: Frequency of Burglary/Theft/Robbery per Location As portrayed on the map above there are properties in Kenosha that experienced more than one crime during the five month period. There were two particular addresses in which there were 26-34 crimes of burglary/theft/robbery in the five month time span. 23 | P a g e There were many locations in which only one crime was committed at that address during the same time period. Figure 8: Frequency of Domestic Violence Crime per Location 24 | P a g e The above map portrays that each location has less than four crimes committed at each address. On two occasions there are three crimes committed at a particular spot, while most locations have one crime committed. Figure 9: Frequency of Drug and Alcohol Crimes per Location The figure above shows that there are a few areas with more than fifteen crimes per location, while there are areas in which there are only a few crimes committed per address. 25 | P a g e Figure 10: Frequency of Assault and Battery Crimes per Location Battery and assault crimes per location are relatively few. There are a few locations in which crime rate is slightly higher in particular locations. 26 | P a g e Figure 11: Frequency of Disorderly Conduct Crimes per Location The figure above shows that there are only a few addresses in which there are a significantly higher number of crimes. There is a cluster of crime that occurs within a few blocks. 27 | P a g e Average Nearest Neighbor Table 1: Average Nearest Neighbor Results Observed Expected Nearest Mean Mean Neighbor Crime Type Distance Distance Ratio z-score p-value Burglary/Theft/Robbery 0.00089 0.002435 0.36556 -27.9685 0 Drug and Alcohol 0.000968 0.003184 0.304122 -24.0366 0 Domestic Violence 0.002794 0.005173 0.540206 -9.26735 0 Battery/Assault 0.001638 0.002983 0.549293 -8.79309 0 Disorderly Conduct 0.000702 0.002492 0.281689 -29.918 0 All 0.000633 0.002114 0.299306 -37.8907 0 The test results provide evidence that crimes are heavily clustered in Kenosha because the observed mean distance is significantly less than the expected mean distance. The z-score is the most important statistic in the above table. If the z-score is less than -2 or -3 then it is evident that clustering did not occur by chance. With values as low as 37.89 to -8.79, it indicates that there is a high level of statistical significance to the clustering of crime events. Burglary/Theft/Robbery, Drug and Alcohol, and Disorderly Conduct appear to have a higher statistical significance of the clustering, than domestic violence and battery/assault do. Hotspot Analysis 28 | P a g e Using the Getis- Ord Gi test it was found that there were different patterns of hotspots throughout. While some crimes had no hotspot apparent, there were others that had many crime pockets. The following figures (Figures 12-13) portray the hotspot analysis of burglary/theft/robbery crimes. Figure 12: Hotspot Analysis of Burglary/Theft/Robbery Crimes 29 | P a g e Figure 13: A Close-up Hotspot Analysis of Burglary/Theft/Robbery Crimes Figure 12 and Figure 13 show that there are hotspots of crime in the neighborhood south of 52nd street, east of Pershing Blvd, and west of Green Bay Road. The map also shows cold spots north of 142. There are also a few more hotspots and cold spots scattered around. 30 | P a g e Figure 14: Hotspot Analysis of Domestic Violence Crimes The figure above shows that there are no hotspots for domestic violence crimes. 31 | P a g e Figure 15: Hotspot Analysis of Drug and Alcohol Crimes 32 | P a g e Figure 16: A Close-up Hotspot Analysis of Drug and Alcohol Crimes Figure 15 and Figure 16 show that there are hotspots. With a 99% confidence level there are many hotspots along 63rd St in between 22nd Ave and Sheridan Road. There are also many cool spots, including along 52nd St, in between 30th Ave and Pershing Blvd. 33 | P a g e Figure 17: Hotspot Analysis of Assault and Battery Crimes The above figure shows that there are 4 areas in which a hotspot occurs, while most areas show no hotspots. 34 | P a g e Figure 18: Hotspot Analysis of Disorderly Conduct Crimes 35 | P a g e Figure 19: A Close-up Hotspot Analysis of Disorderly Conduct Crimes Figures 18 and 19 above show the hotspot analysis of Disorderly Conduct Crimes. There are many hotspots just north of and along 63rd street, south of 60th street, east of 30th avenue, and west of Sheridan Road. 36 | P a g e Demographic Analysis Two crime types that showed clustering and hotspots were disorderly conduct and drug and alcohol crime. These crime types were overlaid and compared to three different demographic variables, including average household income, diversity index, and percent of male population over 18. Figure 20 and 21 below show the relationship between hotspots of disorderly conduct crimes and drug and alcohol crimes and average household income. 37 | P a g e Figure 20: Disorderly Conduct Hotspots VS Average Household Income The figure above shows that in areas of crime, hotspots i.e. clusters of high crime locations, the levels of income are lower. There are a few crimes committed in the highest average income locations; however, there is no hotspot of crime present there. There are a high number of crimes committed in the central portion of Kenosha, and also low income in those areas. 38 | P a g e Figure 21: Drug and Alcohol Hotspots VS Average Household Income The above figure shows that in areas of lower income there are a significantly higher number of crimes committed than in areas of higher income. There are a few outliers, in which crime is committed in a high income location. Drug and Alcohol crime hotspots align well with lower income neighborhoods. 39 | P a g e Figure 22: Disorderly Conduct Hotspots VS Diversity Index 40 | P a g e Figure 23: Drug and Alcohol Hotspots VS Diversity Index In areas where diversity is higher, there is a clustering of higher crime rates for both Disorderly Conduct and Drug and Alcohol Crimes. 41 | P a g e Figure 24: Disorderly Conduct Hotspot VS Percent of Male Population Over 18 The above figure shows that crimes are less concentrated in areas that have a higher percent of males over 18. 42 | P a g e Figure 25: Drug and Alcohol Hotspot VS Percent of Male Population Over 18 The above map shows that high crimes are more clustered in areas that have a lower percent of male population over 18. 43 | P a g e Discussion and Conclusion The Kenosha Police Department can use the findings from this research to focus on certain areas to patrol. The police department only has a limited amount of resources and patrol officers available so knowing the distribution of crime and where certain hotspots are can be a huge help. The police can focus on the more dangerous areas of the city and concentrate its efforts to make sure that Kenosha is a safe place. Kenosha has a lot of room to grow when it comes to management of crimes. Hotspot analysis does tell a lot about crimes. As the data portrayed there were areas in which there were hotspots for certain crimes, while not for others. The crimes that showed hotspots were burglary/theft/robbery, drugs and alcohol, and disorderly conduct. Police can focus on these said areas in order to target high crime areas. Looking at different demographic and social factors can also be a sign of what could affect crimes. As shown in the different figures, in the areas where crime is higher income is significantly lower, diversity is higher, and percentage of males over 18 is lower. The above information can be used to analyze the spatial patterns and trends of crime in Kenosha and be a helpful resource to the police department. Further Research There are numerous factors that could affect crime rates for different crime types in a given area, therefore it is important to analyze crime on a regular basis. While the data analyzed may be sufficient for now, in two years many thing could completely 44 | P a g e change. If crime is analyzed on a regular basis, then the police can look at rates and patterns of crime and compare it to other data sets. Taking a look at more demographic features would be important as well. Certain demographic factors could have an effect on crime rates, therefore studying this can also enable police to make a very educated decision on how to prevent more crimes in their city. The data available was only for a five month time span; therefore crimes that are affected by seasonal change were not analyzed. Since aggressive crimes, such as homicides tend to occur more often in the warmer months, studying the trends of homicides in a five month period is not going to help educate police officers. There are going to be more homicides in the summer months than in the winter, so using hotspots to predict where crimes are more likely to be committed will ultimately not have any merit, due to the fact that rates change between summer months and winter months. The crimes analyzed here have not been known to be affected by seasonal trends, nevertheless, it would be interesting to compare the summer months to other times of the year if data from the entire year is available. If given the data for a whole year at a time, then all crime types should be studied. Only five crime types were analyzed, but there are many other crimes that have not been analyzed, therefore if a larger data set was obtained, repeating this process using all the types of crimes would make the police more knowledgeable on the big picture in Kenosha, Wisconsin. Different demographic factors could be analyzed to better understand the breakdown of crimes. Studying more demographic features and their relationship 45 | P a g e towards crime can be a really beneficial analysis. References 1. Chainey, Spencer, Lisa Tompson, and Sebastian Uhlig. "The Utility Of Hotspot Mapping For Predicting Spatial Patterns Of Crime." Security Journal, 2008. 2. Currie, Elliott. Crime and Punishment in America. New York: Metropolitan Books, 1998. 3. Gartner, Rosemary. "The Victims of Homicide: A Temporal and Cross-National Comparison."American Sociological Review, 1990, 92. 4. Hipp, J. R., P. J. Curran, K. A. Bollen, and D. J. Bauer. "Crimes of Opportunity or Crimes of Emotion? Testing Two Explanations of Seasonal Change in Crime." Social Forces, 2004, 1333-372. 5. Jean, Peter K. B. Pockets of Crime Broken Windows, Collective Efficacy, and the Criminal Point of View. Chicago: University of Chicago Press, 2007. 6. Kramer, R. C. "Poverty, Inequality, And Youth Violence." The Annals of the American Academy of Political and Social Science, 2009, 123-39. 7. Lim, Up, and George Galster. "The Dynamics of Neighborhood Property Crime Rates." The Annals of Regional Science, 2009, 925-45. 8. Messner, Steven F., and Richard Rosenfeld. Crime and the American Dream. 2nd ed. Belmont, CA: Wadsworth Pub., 1997. 9. Quetelet, Lambert A.J. A Treatise on Man: And the Development of His Faculties. 1969. 10. Shihadeh, Edward S., and Darrell J. Steffensmeier. "Economic Inequality, Family Disruption, and Urban Black Violence: Cities as Units of Stratification and Social Control." Social Forces, 1994, 729. 46 | P a g e