1960-2014 UCR violent crimes data

Data obtained from https://www.ucrdatatool.gov

require(ggplot2)
require(dplyr)
#library(plyr)
library(knitr)
library(DT)
library(xtable)
require(ggplot2)
require(reshape)
allc<-read.csv("C:/Users/vivek/Downloads/CrimeStatebyState.csv",header= TRUE)

Sneak peak into the data

##   Year Population Violent.crime.total Murder.and.nonnegligent.Manslaughter
## 1 1960  179323175              288460                                 9110
## 2 1961  182992000              289390                                 8740
## 3 1962  185771000              301510                                 8530
## 4 1963  188483000              316970                                 8640
## 5 1964  191141000              364220                                 9360
## 6 1965  193526000              387390                                 9960
##   Legacy.rape..1 Revised.rape..2 Robbery Aggravated.assault  X X.1 X.2 X.3
## 1          17190              NA  107840             154320 NA  NA  NA  NA
## 2          17220              NA  106670             156760 NA  NA  NA  NA
## 3          17550              NA  110860             164570 NA  NA  NA  NA
## 4          17650              NA  116470             174210 NA  NA  NA  NA
## 5          21420              NA  130390             203050 NA  NA  NA  NA
## 6          23410              NA  138690             215330 NA  NA  NA  NA

##    Year Population Violent.crime.total
## 53 2012  313873685             1217067
## 54 2013  316497531             1199684
## 55 2014  318857056             1197987
## 56 2015  321418820             1197704
## 57   NA         NA                  NA
## 58   NA         NA                  NA
##    Murder.and.nonnegligent.Manslaughter Legacy.rape..1 Revised.rape..2
## 53                                14866          85141              NA
## 54                                14319          82109          113695
## 55                                14249          84041          116645
## 56                                15696             NA              NA
## 57                                   NA             NA              NA
## 58                                   NA             NA              NA
##    Robbery Aggravated.assault  X X.1 X.2 X.3
## 53  355051             762009 NA  NA  NA  NA
## 54  345095             726575 NA  NA  NA  NA
## 55  325802             741291 NA  NA  NA  NA
## 56      NA                 NA NA  NA  NA  NA
## 57      NA                 NA NA  NA  NA  NA
## 58      NA                 NA NA  NA  NA  NA

Creating a copy of the data

#taking out any na values
df<-na.omit(df)

dfcopy<-df

head(dfcopy)

##   Year population total_violent murder_manslaughter
## 1 1960  179323175        288460                9110
## 2 1961  182992000        289390                8740
## 3 1962  185771000        301510                8530
## 4 1963  188483000        316970                8640
## 5 1964  191141000        364220                9360
## 6 1965  193526000        387390                9960

tail(dfcopy)

##    Year population total_violent murder_manslaughter
## 51 2010  309330219       1251248               14722
## 52 2011  311587816       1206031               14661
## 53 2012  313873685       1217067               14866
## 54 2013  316497531       1199684               14319
## 55 2014  318857056       1197987               14249
## 56 2015  321418820       1197704               15696

Please note that data for year 2015 obtained from ‘https://ucr.fbi.gov/crime-in-the-u.s/2015/crime-in-the-u.s.-2015/tables/table-1’ and was appended to the ucr dataset which has data from 1960-2014

Scaling ‘population’ column for better visualization

df$population<-df$population/1000000

Simple plots for first glance

plot(df$Year,df$population, xlab = 'Year', ylab = 'Population (in millions)',main = "1960-2015 US Population" )

plot(df$Year,df$total_violent, xlab = 'Year', ylab = 'Violent crime', main = "1960-2015 All Violent Crimes")

Plotting with qplot

qplot(df$Year,df$population, xlab = 'Year', ylab = 'Population (in millions)',main = "1960-2015 US Population" ,geom = c("point", "smooth"))

## `geom_smooth()` using method = 'loess'

qplot(df$Year,df$total_violent, xlab = 'Year', ylab = 'Violent crime', main = "1960-2015 All Violent Crimes",geom = c("point", "smooth"))

## `geom_smooth()` using method = 'loess'

As can be seen from graph below, crime rates have been reducing significantly over the years.

qplot(df$Year,df$murder_manslaughter, xlab = 'Year', ylab = 'Murder & Non negligent Manslaughter',main = "1960-2015 US Murder & Manslaughter" ,geom = c("point", "smooth"))

## `geom_smooth()` using method = 'loess'

Attempting to plot Population, Total Violent Crimes , Murder & manslaughter(involuntary) in one plot using ggplot2

df$total_violent<-df$total_violent
df$murder_manslaughter<-df$murder_manslaughter


df2 <- melt(data = df, id.vars = "Year")

ggplot(data = df2, aes(x = Year, y = value, colour = variable)) + geom_line()+ggtitle("1960-2015 US Population Violent Crime and Murders") +
  labs(x="Year",y=" Numbers (in millions)")

Above plot isn’t very readable, so let’s scale the columns in the dataframe

df$total_violent<-df$total_violent/10000
df$murder_manslaughter<-df$murder_manslaughter/100
#making sure all values are in same range
df[1:4,]

##   Year population total_violent murder_manslaughter
## 1 1960   179.3232        28.846                91.1
## 2 1961   182.9920        28.939                87.4
## 3 1962   185.7710        30.151                85.3
## 4 1963   188.4830        31.697                86.4

df2 <- melt(data = df, id.vars = "Year")

Plotting scaled values

Please take note of the following measurement units for the above graph:

Population is being measured in the millions (1 unit on graph = 1 million)
Total Violent crime is being measured in the 10,000’s (1 unit on graph = 10000)
Murder and Manslaughter(non-negligent) is being measured in the 100’s (1 unit on graph = 100)

Clearly crime rates have been reducing over the years.

dfcopy <- dfcopy[order(-(dfcopy$murder_manslaughter)),]
aa<-1:nrow(dfcopy)
rownames(dfcopy)<-aa
dfcopy$Rank<-rownames(dfcopy)

Ranking the years in descending order of murders and involuntary manslaughter

kable(dfcopy)

Year	population	total_violent	murder_manslaughter	Rank
1991	252153092	1911767	24703	1
1993	257782608	1926017	24526	2
1992	255029699	1932274	23760	3
1990	249464396	1820127	23438	4
1994	260327021	1857670	23326	5
1980	225349264	1344520	23040	6
1981	229465714	1361820	22520	7
1995	262803276	1798792	21606	8
1989	246819230	1646037	21500	9
1979	220099000	1208030	21460	10
1982	231664458	1322390	21010	11
1974	211392000	974720	20710	12
1988	244498982	1566221	20675	13
1986	240132887	1489169	20613	14
1975	213124000	1039710	20510	15
1987	242288918	1483999	20096	16
1996	265228572	1688540	19645	17
1973	209851000	875910	19640	18
1978	218059000	1085550	19560	19
1983	233791994	1258087	19308	20
1977	216332000	1029580	19120	21
1985	237923795	1327767	18976	22
1976	214659000	1004210	18780	23
1984	235824902	1273282	18692	24
1972	208230000	834900	18670	25
1997	267783607	1636096	18208	26
1971	206212000	816500	17780	27
2006	299398484	1435123	17309	28
2007	301621157	1422970	17128	29
1998	270248003	1533887	16974	30
2005	296507061	1390745	16740	31
2003	290788976	1383676	16528	32
2008	304059724	1394461	16465	33
2002	287973924	1423677	16229	34
2004	293656842	1360088	16148	35
2001	285317559	1439480	16037	36
1970	203235298	738820	16000	37
2015	321418820	1197704	15696	38
2000	281421906	1425486	15586	39
1999	272690813	1426044	15522	40
2009	307006550	1325896	15399	41
2012	313873685	1217067	14866	42
1969	201385000	661870	14760	43
2010	309330219	1251248	14722	44
2011	311587816	1206031	14661	45
2013	316497531	1199684	14319	46
2014	318857056	1197987	14249	47
1968	199399000	595010	13800	48
1967	197457000	499930	12240	49
1966	195576000	430180	11040	50
1965	193526000	387390	9960	51
1964	191141000	364220	9360	52
1960	179323175	288460	9110	53
1961	182992000	289390	8740	54
1963	188483000	316970	8640	55
1962	185771000	301510	8530	56

In terms of murder and manslaughter , 2014 and 2015 are no where near the bloodiest of years, in fact they are ranked #38 and #47 respectively in the 55 years between 1960-2015 But 2015 definitely saw a small rise in the murder rate.

The availability of 2016 2017 data could help us decide whether or not the violent crime rate has risen .

https://www.linkedin.com/in/vivekmangipudi/

US Population and Violent Crimes 1960-Present

Vivek Mangipudi

8 March 2017

1960-2014 UCR violent crimes data

Data obtained from https://www.ucrdatatool.gov

Sneak peak into the data

Creating a copy of the data

Please note that data for year 2015 obtained from ‘https://ucr.fbi.gov/crime-in-the-u.s/2015/crime-in-the-u.s.-2015/tables/table-1’ and was appended to the ucr dataset which has data from 1960-2014

Scaling ‘population’ column for better visualization

Simple plots for first glance

Plotting with qplot

As can be seen from graph below, crime rates have been reducing significantly over the years.

Attempting to plot Population, Total Violent Crimes , Murder & manslaughter(involuntary) in one plot using ggplot2

Above plot isn’t very readable, so let’s scale the columns in the dataframe

Plotting scaled values

Please take note of the following measurement units for the above graph:

Clearly crime rates have been reducing over the years.

Ranking the years in descending order of murders and involuntary manslaughter

In terms of murder and manslaughter , 2014 and 2015 are no where near the bloodiest of years, in fact they are ranked #38 and #47 respectively in the 55 years between 1960-2015 But 2015 definitely saw a small rise in the murder rate.

The availability of 2016 2017 data could help us decide whether or not the violent crime rate has risen .