FBI Gun Data , US Census Data and Kaggle Gun Violence Data are three independent tables. Their common variables include states and it requires data cleaning with removal of nan values and removal of few columns for getting accurate insights.
Tip: In this section of the report, brief introduction to the dataset you've selected for analysis. At the end of this section, described the questions that I plan on exploring over the course of the report.
And then I familiarize myself with the variables and the dataset context for ideas of what to explore.
I referred to below website for better understanding the data metrics and statistical outcomes I'll be working with.
FBI Gun Data , US Census Data and Kaggle Gun Violence Data are three independent tables. Their common variables include states and it requires data cleaning with removal of nan values and removal of few columns for getting accurate insights.
Find corelation between the trend of casualities in mass shootings vs Total Count by states from 2015 to 2016 ? What census data is most associated with high gun per capita? Which states have had the highest growth in gun registrations? What is the overall trend of gun purchases?
Missing or nan values in the dataset which could affect our understanding of the long term insights of particular variable. Presence of outliers in the dataset adversely affects strength of other data values in calculations. Non-standardisation of the datasets
Kentucky has the largest count of total gun sales and one of the lowest count total casualities in the mass shootings between 2015 and 2016 which suggests presence of tougher gun permit verifications and indicates more rigorous checking of permit licences. On the other hand Florida has the highest count of casualities which calls for better system checks in the state and better background checks to prevent such incidents in the future.
There are few weak associations between the census variables and gun per capita: The positive association between gun per capita and variables which includes: White alone, percent, July 1, 2016, (V2016) Persons 65 years and over, percent, April 1, 2010 owner-occupied housing unit rate, 2011 -2015 The negative association between gun per capita and variables which includes: 2011-2015 Asian alone, percent, July 1, 2016, (V2016) Foreign born persons, percent, 2011-2015 Median gross rent, 2011-2015
Count of gun registrations have gone up in every state except Utah inferred from positive growth in the gun registrations. Alaska has the biggest growth in gun registrations by more than 400% followed by Wyoming, Montana, Kansas and Arkansas. This could suggest either they are outliers or , increased dependence and trust of people on firearms and decreased trust on police institutions or increased safety concers caused by various number of reasons.
From the Line chart of gun sales vs years 1997 - 2016, there is increasing trend of gun purchases with sudden increases in year 2015 and 1998 and decrease in year 2016 which is partially due to data collection of only 9 months in that year.
There is postive trend in gun purchase over the years and state is the most important variable for comparing the data along with few weak association with the census variables.