I. Summary

The csv file used shows a summary of Arrests based on Violent Crimes in the United States. This data set contains statistics, in arrests per 100,000 residents for assault, murder, and rape in each of the 50 US states in 1973. Also given is the percent of the population living in urban areas.
Data Exploration will highlight summary information about the data.
Data Wrangling will be used specifically to rename and add columns that would be beneficial to this investigation.
Graphics will display the basic graph plots: Scatter Diagram, Box Plot, and Histogram. Additionally, using ggplot2, Bar Chart, Stack Chart, and Pie Chart wil be displayed.
Question - In 1973, as Average Income and Education Expenditure Per Student increases in each State, does Total Arrests due to Violent Crimes decrease?

II. Data Exploration

Data exploration activities satisfied in this section are:

BONUS Question - data was derived directly from a raw git hub url: https://raw.githubusercontent.com/RommyGraphs/MSDA/master/2018Workshop/USArrests.csv
Summary function is used to include information about appropriate mean, median, quartiles, and any other relevant information about the data set.
Summary Conclusions - brief analysis of the summary data.

A. Summary Data

##           X          Murder          Assault         UrbanPop    
##  Alabama   : 1   Min.   : 0.800   Min.   : 45.0   Min.   :32.00  
##  Alaska    : 1   1st Qu.: 4.075   1st Qu.:109.0   1st Qu.:54.50  
##  Arizona   : 1   Median : 7.250   Median :159.0   Median :66.00  
##  Arkansas  : 1   Mean   : 7.788   Mean   :170.8   Mean   :65.54  
##  California: 1   3rd Qu.:11.250   3rd Qu.:249.0   3rd Qu.:77.75  
##  Colorado  : 1   Max.   :17.400   Max.   :337.0   Max.   :91.00  
##  (Other)   :44                                                   
##       Rape      
##  Min.   : 7.30  
##  1st Qu.:15.07  
##  Median :20.10  
##  Mean   :21.23  
##  3rd Qu.:26.18  
##  Max.   :46.00  
##

B. Summary Conclusions:

Murder - For the entire USA in 1973, the mean murder arrests is 7.78 per 100,000 residents. Respectively, the Median, 1st Quartile, 3rd Quartile, Minimum, and Maximum rates are 7.25, 4.08, 11.25, 0.8, and 17.4 murder arrests per 100,000 residents.
Assault - For the entire USA in 1973, the mean assault arrests is 170.8 per 100,000 residents. Respectively, the Median, 1st Quartile, 3rd Quartile, Minimum, and Maximum rates are 159, 109, 249, 45, and 337 assault arrests per 100,000 residents.
Rape - For the entire USA in 1973, the mean rape arrests is 21.23 per 100,000 residents. Respectively, the Median, 1st Quartile, 3rd Quartile, Minimum, and Maximum rates are 20.1, 15.07, 26.18, 7.3, and 46 rape arrests per 100,000 residents.
Urban Population - For the entire USA in 1973, the mean percentage of urban population was 65.54%. Respectively, the Median, 1st Quartile, 3rd Quartile, Minimum, and Maximum rates are 66, 54.50, 77.75, 32, and 91 percents of Urban Population.

At this time, the summary information does not reveal any significant information about the US Arrests based on Violent Crimes in 1973. More detailed analysis should be investigated.

III. Data Wrangling

Data wrangling activities satisfied in this section are:

Column Additions - Add three new columns: that will store Average Income, Average Education Expense per Person, and Violent Crimes. The defaults will be blank.
Column Renaming - Some column renaming to make it more readable to users. For example title for State is blank initially. It was changed to State for user readability. UrbanPop was renamed to PctUrbanPop.
** Assigning Values** - The first two new columns will be assigned values dervied from the Census data. Violent Crimes column will be assigned the sum of Murder, Assault, and Rape arrests for that State.
Summary function is used to include information about appropriate mean, median, quartiles, and any other relevant information about the NEW data set. This is really a test function to verify the data was shown.

Note: I have added Income and Education Expenditure columns from the Census website.Please note that Income is based on 1972 information, therefore, data trends may be inaccurate for this discussion. https://www.census.gov/library/publications/1973/compendia/statab/94ed.html

Summary Information of New Data Set

##         State        Murder          Assault       PctUrbanPop   
##  Alabama   : 1   Min.   : 0.800   Min.   : 45.0   Min.   :32.00  
##  Alaska    : 1   1st Qu.: 4.075   1st Qu.:109.0   1st Qu.:54.50  
##  Arizona   : 1   Median : 7.250   Median :159.0   Median :66.00  
##  Arkansas  : 1   Mean   : 7.788   Mean   :170.8   Mean   :65.54  
##  California: 1   3rd Qu.:11.250   3rd Qu.:249.0   3rd Qu.:77.75  
##  Colorado  : 1   Max.   :17.400   Max.   :337.0   Max.   :91.00  
##  (Other)   :44                                                   
##       Rape         AvgIncome      AvgEducExp     ViolentCrimes  
##  Min.   : 7.30   Min.   :3100   Min.   : 590.0   Min.   : 53.1  
##  1st Qu.:15.07   1st Qu.:3700   1st Qu.: 804.5   1st Qu.:128.2  
##  Median :20.10   Median :4300   Median : 953.0   Median :185.2  
##  Mean   :21.23   Mean   :4228   Mean   : 972.7   Mean   :199.8  
##  3rd Qu.:26.18   3rd Qu.:4575   3rd Qu.:1116.8   3rd Qu.:285.8  
##  Max.   :46.00   Max.   :5300   Max.   :1584.0   Max.   :382.3  
##

IV. Graphics

Scatter Plot

In this experiment, I will use the columns Violent Crimes and Average Education Expense to see if there is any correlation.

Reference: https://www.tutorialspoint.com/r/r_scatterplots.htm

Box Plot

In this experiment, I will use columns Murder and Average Income Greater than $3500 to see if there is any correlation.

Reference 1: https://www.r-bloggers.com/5-ways-to-subset-a-data-frame-in-r/
Reference 2: http://www.theanalysisfactor.com/r-tutorial-part-13/

Histogram

In this experiment, I will use Violent Crimes column to make a basic histogram.

Reference: https://www.statmethods.net/graphs/density.html

Bar Chart

In this experiment, I will use data from Violent Crimes vs. Average Income Greater than 5000 to develop a simple bar chart. I will also check for a correlation.

Stack Chart and Pie Chart

In this experiment, we will use the bar chart data to create a simple Stack Chart and Pie Chart.

V. Conclusions About Project Based on Data Experiments Conducted

Question - In 1973, as Average Income and Education Expenditure Per Student increases in each State, does Total Arrests due to Violent Crimes decrease?

Based on the available data given, conclusively, there is no correlation between Average Income and Education Expenditure per Student in each State vs. Total Arrests due to Violent Crimes in 1973. The Data Exploration portion of the Project only showed means for Arrests due to Murder, Rape, and Assaults. On average, indicents of Murder Arrests were less than Rape Arrest and incidents of both Murder and Rape Arrests were significantly less than Assault Arrests. The Data Wrangling portion of the project introduced two new variables for each State: Average Income and Average Education Expenditure for Each Student in US Dollars. The summary data from this new table did not show any significant trends that would answer the question.

Things get interesting in the Graphics portion of the Project. The Scatter Plot diagram highlights Arrests from Violent Crimes vs the Average Education Expense per Student by State. Violent Crimes is the sum of all Murder, Rape, and Assault Arrests for each State. I was hoping that as Average Income and Average Education Expenditure increased, the number of Arrests of Violent Crimes would decrease. Unfortunately, the Scatter Plot chart tells a different story regardless if it is Average Education Expense or Average Income. There are no Violent Crimes arrests with Average Education Expense less than 500 dollars and there are no Violent Crime arrests with Average Income less than 3000 dollars. The Box Plot chart showing the relationship between Average Income Greater than 3500 dollars and Murder Arrests is even more disturbing. There is no inverse relationship and the data points are all over the map. The Bar Chart shows 6 States with Average Incomes greater than 5000 dollars AND not one state has Violent Crime arrests less than 100. The Histogram, Pie Chart, and Stack Chart do not show any information to indicate any correlation.

In conclusion, based on the data given, increase in Average Income and Education Expenditure Per Student in each State does not indicate a decrease in Arrests due to Violent Crime.

MSDS Workshop 2018 - R Final Project

Romerl Elizes

January 16, 2018