Chapter 4 Missing Values
In this chapter , we take a deep look into the data that we have and analyze missing data patterns, furthermore we try to explain the reasoning behind the missing data.
4.1 Missing Data in Juvenile Arrest Records
We take a look at the Juvenile Arrest Records data , in the heat map below , we can see that the data is missing for “Running Away” and the “Forcible Rape” Offenses from years after 2009.
With the knowledge that there are few missing data from the Year 2010 to Year 2019 , we take a look at the precentage of the data that are missing in the graph.
Looking into the literature of the Juvenile Arrests Data , we see that he FBI no longer reports arrests for running away since 2010 , this is why we see missing values in the graph.
In similar terms, in 2013, the FBI broadened the definition of rape, removing the phrase “forcible” from the offense name and description , this led to difference in reporting practices of the different reporting agencies,this means that we do not have accurate data for the offense after 2012.
4.2 Missing Data in Gun Violence Archive
The Stacked bar Plot for the Gun Violence Dataset shows that there are no missing data , this goes to show that Gun violence is being tracked meticously and all the information is duly available.
As there is no missing data in the gun violence archive data , there is no inference of reason that can be drawn from it.
4.3 Missing Data in Hate Crime Data
In the FBI Uniform Crime Reporting Program , we have multiple datasets where in each dataset shows the HateCrime data against one facet.
4.3.1 Hate Crime data by Offense
Due to the change of definition of by FBI as described above , we see the same missing data pattern in the Uniform Crime Reporting data.