Introducing 

Prezi AI.

Your new presentation assistant.

Refine, enhance, and tailor your content, source relevant images, and edit visuals quicker than ever before.

Loading content…
Loading…
Transcript

violent crimes and college graduation rates

Oriana Giusti Rincon

This analysis contains information about the rate of college graduates (in %) and the number of violent crimes committed per 100,000 people of the population in each of the 50 states

intro

I expect to see a negative relationship because People who don’t receive an education are usually in an unstable environment and aren’t surrounded by people who can teach them the importance of morals and values. Also, the more education a person gets, the more they develop patience, awareness for the effect of consequences, care for their future and less available free time. As college graduate rate increases I think violent crime rate will decrease.

scatterplot

scatterplot of data

The scatterplot appears to show a moderate linear negative relationship between college graduates’ rate (explanatory variable) and the violent crime rate per 100,000 of the population (response variable). I do suspect of some outliers.

residual plot

The residual plot above proves that there are possible outliers in our data. Those are: Alaska at (30, 437.9294), New Mexico at (27, 341.317), Maine at (32, -266.529).

possible outliers

why are they outliers

I struggled to find a resemblance in the 3 outliers but at least 2 of them, the ones with higher crime rates might signal an explanation since they are border states, New Mexico, and Alaska. Maybe the explanation for such high crime rate is their likeliness to have a higher illegal migration of people that in many cases come to this country in a vulnerable condition making them prone to commit violent acts.

Alaska and New Mexico

Why is Alaska/ New Mexico an outlier

High outliers

the crime rate in both of these states is higher

Alaska

- long distances, many towns and many communities located in remote areas making it difficult for law enforcement to protect people

- high alcohol beverage intake

- weather conditions

- long nights

New Mexico

- drug trafficking.

Why is it an outlier?

Maine

Maine unlike Alaska and New Mexico is an outlier due to its low crime rate. After analyzing, I found that Maine has mostly an older and homogenous population consisting of Anglo Saxons. This could contribute to the low violent crime rate since there are barely any racial and cultural differences that cause conflict among people

LSRL:

: ŷ(violent crime rate per 100,000 of the pop.) = 770.9948 - 12.3708x (college graduates’ rate)

LSRL, R and slope

r value

The correlation value of this scatterplot is -0.4165. This number further proves the relationship between these two variables to be negative, linear and moderate.

r squared

The coefficient of determination is 0.1735. What this means is that 17.35% of the variation in the violent crime rate per 100,000 of the population of each state can be explained by the variation in college graduation rates in each state.

slope

The slope is -12.3708. This means for every increase in college graduation rates by 1%, the violent crime rate decreases by 12.3708 on average.

linear regression t - test

Is there a linear relationship between the rate of violent crime rate and college graduation rates per state? I will be testing at a significance level of .1 and .05.

Ho: there is no linear relationship between the violent crime rate and college graduation rate per state. (B = 0)

Ha: There is a linear relationship between the violent crime rate and college graduation rate per state. (B≠ 0)

Assumptions necessary for inference were judged reasonable.

linear reg t-test

conclusion

conclusions

p-value = 0.0026

Since the p-value is lower than alpha at both 10% and 5% significance levels (0.0026 ≤ 0.05 ) (0.00026 ≤ 0.1) then we reject the null hypothesis. There is sufficient evidence to prove that there is a linear relationship between the violent crime and college graduation rate per state. Since the slope is -.12.3708, we also know that this relationship is negative.

final thoughts

After doing this analysis, I have concluded that there is a negative linear relationship between the violent crime rate and college graduates’ rate for the 50 states. However, there are some variables that could affect this as well which I didn't account for, like the population of each state, how close they are to the border, whether they are in the East or West coast, the economy, poverty etc. There appears to be a linear negative trend between both variables. I believe this relationship happens because the different awareness and responsibility skills education grants people is lacked by those who don’t graduate college. However because of my low coefficient of determination, I wouldn't say college graduation rates are good for predicting violent crime rate.

Learn more about creating dynamic, engaging presentations with Prezi