The United States of America is one of the only few countries that constitutionally protects the right to bear arms. However, people in politics and the news have debated if there should be restrictions on what types or how many guns people can have in an effort to curb gun violence. Thus, when presented with this research project, we wanted to educate ourselves on this topic and look into what factors potentially contribute to gun violence in the United States.
The question that we based our research project on was what factors could potentially lead to an increase or decrease in a state’s firearm mortality rate? However, before we could answer this question, we had to decide on some variables that we thought would have an significant impact on our response variable. In the end, we chose to focus on five variables. They were the number of gun laws, gun ownership rates, poverty rates, urban population percent, and the state’s political party. Our response variable was the firearm mortality rate, which is the number of all firearm deaths per state, both unintentional and intentional, per 100,000 people. When we began looking for data, we first explored Kaggle. However, there wasn’t a dataset that had the variables we wanted to research. Thus, we had to gather the different variables through different sources, which was a challenge for us. In order to avoid biased data, we individually gathered firearm data state by state though the CDC and Census Bureau.
Our data contained four quantitative variables and one categorical variable. Our first quantitative variable was the number of gun laws. We hypothesized that a increase in the number of restrictions and regulations for guns would decrease the firearm mortality rate. For our second quantitative variable, gun ownership rate, we hypothesized that an increase in amount of guns people have would result in an increase in gun violence. Our third and fourth quantitative variables were the state’s poverty rate and urban population percent. Our prediction for the poverty rate was higher the rate, the higher the firearm mortality rate. This is because those who are better off financially will move away out of places with high rates of violence leaving those who are financially unable to. For the percent of urban population, we predicted that a state with higher percent would have more violence, hence, a higher percent of gun violence. The last variable we choose was the 2016 political party of a state. We were not sure what to predict for this categorical variable, but we thought it would be interesting to see if there was a relationship between this and the response variable.
After we choose our predictors, we first checked for multicollinearity. Multicollinearity exists when two or more of our independent variables are highly correlated with each other. This is a problem if it exists because it can skew the results of our research. For example, multicollinearity can inflate our standard errors of our coefficients, which means that the results could potentially tell us that a variable is insignificant when it truly is.
To test this, we used the VIF test in R Studio. In order to rule out multicollinearity, our VIF factor output would have to be less than ten.
A professional writer will make a clear, mistake-free paper for you!Get help with your assigment
Please check your inbox