Study of Seat Belt Usage
- Griffen Herrera
- Feb 27, 2023
- 6 min read
This dataset focuses on the effects of seatbelt law had on the United States during the years 1983-1997. This dataset came from Cohen, A.and Einav, L. (2003). “The Effects of Mandatory SeatBelt Laws on Driving Behavior and Traffic Fatalities”. This dataset has a total of 765 observations. There are a total of 11 variables, which 6 variables are categorical and 5 variables are numeric. The 6 categorical variables consist of state, year, speed limit (Over 70 mph or not), legal drinking age (21+ or not), blood alcohol limit (Less than 0.08 or other limit), and enforcement of seat belt law. The 5 numerical variables consist of millions of traffic millions per year, fatality rate (Number of fatalities per million of traffic miles), seat belt usage rate, income, and age. Below shows all the different variables in this dataset.

Below shows a partial dataset of New York state's numeric and characteristically variables. It is interesting to see the progression of seatbelt usage from 1984-1997 it increased by 58%. The reason this happened is because the seat belt law started to be enforced in 1985. It is also interesting to see the progression of increasing millions of traffic miles per year and decrease of fatality rate (dropped from 2.5% to 1.4%).

The dataset below shows the year that had the highest fatality rate (Number of fatalities per million of traffic miles) for each state. According to the dataset the highest fatality rate belongs to Alaska and New Mexico (4.5%) in 1983. The lowest of these fatality rate highs is New Jersey with 1.9% in 1986.

The dataset below shows the percentage of seatbelt usage in California, Colorado, and Connecticut from 1990 to 1997. The state of California started to enforce the seatbelt law in 1986 and we can see a steady increase of seat belt usage from 1990-1997. The state of Colorado started to enforce the seatbelt law in 1987 and we can also see a steady increase but at a slower rate than California. Colorado did have a slight decrease from 1991 to 1992, but continued to increase afterwards. The state of Connecticut started to enforce the seatbelt law in 1985 and we can see a steady increase of seat belt usage until 1996 where it dropped down by 10 percent. The cause of the drop could have been due to a car crash fatality which was blamed on the seatbelt (Hartford Courant, 1996).

The table below displays the descriptive statistics of income. It is shown that the maximum income is $35,863 and the minimum income is $8,372 with an average income of about $18,000. The income median is $17,624 with a standard deviation of 4,811.46.

The table below displays the descriptive statistics of millions of traffic miles per year. It is shown that the maximum is 285,612 miles and the minimum is 3,099 miles with an average of about 41,500 miles. The median is 30,319 miles with a standard deviation of 43,961.99.

The table below displays the descriptive statistics of age. It is shown that the maximum age is 39.2 years old and the minimum age is 28.2 years old with an average age of 35.1 years old. The median age is 35.4 years old and a standard deviation of 1.698.

The table below displays the fatality rate per millions of traffic miles in the United States from 1983-1997. The maximum fatality rate is 4.55% and the minimum fatality rate is 0.83% with an average of 2.15%. The median fatality rate is 2.12% with a standard deviation of 0.00617.

The table below displays the seat belt usage in the United States from 1983-1997. Since various states did not enforce the usage of seat belts based on state laws that is why this table has less total observations (556 observations instead of 765 observations). The maximum seat belt usage is 87% and the minimum seat belt usage is 6% with an average of 52.89%. The median seat belt usage is 55% with a standard deviation of 0.1702.

The table below shows the descriptive statistics of income, age, millions of traffic miles per year, fatality rate per millions of traffic miles, and seat belt usage from the two speed limit variables (70 or more speed limit and any other speed limit under 70). While comparing the averages of each numeric variable we can see that income, age, millions of traffic miles per year, and seat belt usage are greater when the speed limit is over 70, but the fatality rate is greater when the speed limit is under 70.

The table below shows the descriptive statistics of income, age, millions of traffic miles per year, fatality rate per millions of traffic miles, and seat belt usage from the two seat belt law variables (Yes - seat belt law is enforced and No - seat belt law is not enforced). While comparing the averages it is obvious that the seat belt usage average will increase significantly, for example when the seat belt law was not enforced the usage was only about 50% but when it was enforced the usage was 70% (seat belt usage went up by 20%).

The table below shows the descriptive statistics of income, age, millions of traffic miles per year, fatality rate per millions of traffic miles, and seat belt usage from the two alcohol limit variables (The blood alcohol limit is less than or equal to 0.08 or something different). While comparing the averages by each variable of alcohol limit, the seat belt usage is greater by 11.4% when the blood alcohol limit is less than or equal to 0.08.

The table below shows the descriptive statistics of income, age, millions of traffic miles per year, fatality rate per millions of traffic miles, and seat belt usage from the two legal alcohol age variables (No - the alcohol age is not 21 years old, Yes - the alcohol age is 21 years old). While comparing the averages by each variable of alcohol age, we can see that the seat belt usage is about 34% greater when the legal age for alcohol consumption is 21 years old.

The table below shows the descriptive statistics of income, age, millions of traffic miles per year, fatality rate per millions of traffic miles, and seat belt usage from three of the fifty-one different variables (Selected Alaska, Georgia, and Kentucky). While comparing the averages by these three states, we can see that Alaska has the greatest average of seat belt usage of 65% usage, then Georgia follows with 46% usage, and lastly there is Kentucky with 32% usage.

The graphs below shows the distribution of per person income. It is apparent that the majority of the per person income is between $13,000 to about $23,000, which is trivial since the average income is about $18,000. Similarly this density plot shows per person income, but shows more of an accurate distribution of percentages. On this plot we can see that the majority of people have an income from $13,000 to $23,000 like said before for the histogram.


The graphs below shows the distribution of millions of traffic miles per year and we can see that the majority of the percentage is between 0-40,000 miles (About 74%). The rest of the distribution varies from 40,000 to 280,000 miles. Based on the density plot over 30% of 10,000 millions of traffic miles per year and afterwards the millions of traffic miles exponentially decays.


The graphs below shows the distribution of age and can see the majority of the people are in their 30s (mostly 34-36 years old). Based on the density plot we can see that almost 25% of the people are 35.8 years old and the majority of people are in their mid-30s.


The graphs below shows the distribution of fatality rate (per year for every state), and based on the graphs we can see that majority the fatality rates is close to 2%. Based on the density plot we can see that the majority of the fatality rate from every state from 1983-1997 ranges from 1.3% to 2.4%.


Lastly the graphs below shows the distribution of seat belt usage from the entire United States during 1983-1997. We can see that the majority of seat belt usage is from 51%-69%. Based on the density plot we can see that it is very similar to the histogram where the majority is from 51%-69%.


Comments