Chibuike Loves Math

This week, I have been looking at the tags column in the American dataset. This column seems to contain additional data about each demonstration in the dataset. This data includes whether or not the protesters were armed, which is interesting because I found that some demonstrations that were labelled as peaceful protests still had armed protesters which was interesting.

The tags column also contained whether or not the protest was a counter demonstration or not and the size of the crowd. I used the str_detect() function in R’s stringr library to extract the information and put them into extra columns.

The crowd size will need additional formatting because the data was represented in a very inconsistent format

This week I kept looking into the outlier police stations that had shooting counts that were 3 standard deviations above the mean. Among these police stations I am looking at how many of them are using body cams. I found out that 77% of the outliers did not use body cams. A question I want to ask is “Are body cams being utilized enough to prevent fatal shootings by police officers?”

I am currently investigating the cities containing the police stations that have shot the most people and seeing if there is any correlation there

Going off of Professor Davis’ work where he identified the few police stations that were involved in 80% of the shootings, I plan to analyze the cities that these police stations are located in to see how the shootings were distributed across the cities

From my explorations on mental illness cases I mentioned in my last post, here are my results:

Fatal Police Shootings Where The Victim Was Suffering A Mental Illness During The Altercation

Black Males – 16.49%

Black Females – 1.28%

White Males – 57.17%

White Females – 4.92%

Asian Males – 2.25%

Asian Females – 0.32%

Hispanic Males – 14.99%

Hispanic Females – 0.54%

 

White males are the most overrepresented group in the dataset, making up over half the population. But when you look proportionally to the population of the United States, Black and Hispanic males also become overrepresented.

 

 

Proportional Representation

Black Males

 16.49%/6% x 100 = 274.8333 %

 

White Males

57.17%/30% = 190.5667% 

 

Hispanic Males

14.99%/8% x 100 = 187.375 %

 

Method

To do this I used the sub setting function, combined with logical operations to capture a subset of each demographic

Black men with mental illness have been historically mistreated in the United States and I have been investigating if there is any evidence of that in the dataset. To do this I am checking if black men with mental illness are being disproportionately represented in the dataset, when compared with the total population of the United States.

I am also looking at the “threat_type” variable compared with the “armed_with” variable. If a person is threatening a policeman by pointing at them, but they are only armed with a blunt weapon, that is not a valid reason to use lethal force and if this happens repeatedly, then it is a problem that should be recorded/reported.

 

 

This week, we are working on a dataset compiled by the Washington Post about fatal police shootings in the United States. This dataset contains information like the name of the victim, their location, the police department responsible for the shooting and more. Using this dataset, our goal for the week is to pose a question about the data and use the dataset to answer that question. The answer of that question will then lead to more questions. I chose to investigate what state had the most frequent shootings and found that it was California. There are further questions this poses, like :

a.) What race of people were most frequently shot by police in California?

b.) What police department in California was involved in the most shootings?

c.) What percentage of people shot by police departments in California were armed?

I will be exploring these questions and more in the coming weeks.

Welcome to UMassD WordPress. This is your first post. Edit or delete it, then start blogging!