Car Accidents Changes a Person’s Life Essay Example

25 percent. It is common to find rumble strips on many good, safe roads. Consequently, safe roads often result in fewer accidents since they are less hazardous for drivers compared to fair, mediocre, or poor roads. Thus, the expected coefficient for this independent variable should be negative. The independent variable SAF represents the funding allocated for highway safety programs per registered motor vehicle in 2003. It is measured by the total federal funds allotted for safety programs in each of the fifty states (in thousands of dollars). These funds are then divided by the total motor vehicle registrations in each state (in thousands of registered drivers). According to Dorn and Barker (2004), drivers who undergo professional highway safety training are safer than those who do not participate in such programs.

Increasing expenditure per registered motor vehicle for highway safety programs results in a reduction of fatal car accidents. This is shown by the negative coefficient associated with the independent variable, MIL, which represents the average total vehicle miles traveled per registered motor vehicle in each state (in thousands) in 2003.

As the mileage of a driver increases, the likelihood of their involvement in a fatal car accident also increases due to spending significant time on the road. Therefore, we can expect a positive coefficient for this independent variable.

The independent variable GAS represents the average price of unleaded fuel in each state in 2003 (in dollars). Economic theory suggests that an increase in gas prices leads to a decrease in gasoline demand. This decrease results in fewer miles being driven and subsequently, there is also a decrease in fatal accidents because fewer individuals choose to drive when gas

prices are high and instead opt for alternative transportation methods.

According to Navon (2001), a study in 2003 found that high speeds on urban interstate roads increase the likelihood of fatal car accidents. This is evidenced by higher crash rates, injury rates, and a greater chance of drivers losing control when traveling at higher speeds. As a result, an increased speed limit is associated with a higher probability of fatal car accidents. Therefore, it is anticipated that the coefficient for the independent variable SPD will be negative.

The variable SBT represents the fine amount for non-compliance with seatbelt regulations in each state in 2003. Imposing fines on drivers for not wearing seatbelts serves as an incentive for them to exercise caution in future instances. Higher fines associated with non-compliance encourage drivers to consistently wear seatbelts, as they try to avoid receiving a substantial penalty. Scientific evidence supports the crucial role of seatbelts in saving lives. According to Robertson (1976), individuals restrained by a seatbelt experience a 50-80% decrease in the chances of death during accidents. Wearing a seatbelt significantly reduces the probability of experiencing a fatal accident.

Therefore, the coefficient of this independent variable, DRIY (representing the percentage of motor vehicle drivers under the age of twenty-five in each state), is expected to be negative. Younger drivers lack experience and may not be familiar with dangerous road conditions. In addition, many young drivers tend to disregard speed limits and often drive faster than what is mandated by the state. According to Bingham and Shope, individuals under 35 years old are at a higher risk of dying in motor vehicle crashes, which further supports the negative correlation between DRIY

and the coefficient. Moreover, young drivers have a higher likelihood of engaging in drug and alcohol abuse.

Based on the arguments mentioned above, it is expected that the independent variable, DRIS, will have a positive sign. DRIS represents the percentage of motor vehicle drivers over the age of sixty-five in each of the fifty states. Elderly individuals often face health issues such as glaucoma or hearing loss that can affect their driving abilities. According to West, Gildengorin, et al (2003), poor vision is the most common impairment among senior drivers. Additionally, reflex skills and certain motor skills of older adults are not comparable to those of younger individuals. Therefore, this independent variable is anticipated to be positively associated with the given factors.

The variable SUV represents the proportion of sport utility vehicle registrations to total vehicle registrations in each state. Despite their popularity, SUVs have a higher risk of rollovers. According to Rivara, Cummings, and Mock (2003), 60% of all rollover accidents occur in SUVs. Comparatively, minivans, trucks, and small cars generally have more safety features than some SUVs. Therefore, it is expected that the coefficient of this variable will be positive because an increased chance of a rollover also increases the likelihood of a fatality.

The variable DPM represents the density of licensed drivers per square mile in 2003. It is calculated by dividing the total number of licensed drivers per state by the number of square miles per state.

In the northeast, states are usually densely populated which leads to a significant number of drivers in small areas. This large concentration of drivers in close proximity adds to the likelihood of accidents due to heavy traffic and

the abundance of vehicles on the road. Consequently, it is anticipated that the coefficient for this independent variable will be positive since having more cars per square mile raises the risk of fatal accidents. It might be unexpected that West Virginia obtains the highest level of federal highway safety program funding because one would typically assume that larger states such as Texas or California would receive greater funding based on their size.

When calculating these values, the focus is on the dollar amount per registered driver. It is surprising that nearly one quarter of West Virginia's registered vehicles have drivers over the age of 65. This is unexpected because Florida is usually associated with having a large elderly population. One possible explanation for Florida not having the highest percentage could be that married senior citizens may only possess one car.

The percentages were calculated based on the count of registered vehicles. Hence, even though a vehicle may have only one registered owner, it can be driven by two individuals.

III. Assessment of Multicollinearity

Multicollinearity occurs when two or more independent variables demonstrate a linear relationship or correlation with each other. Multicollinearity has two important implications. Firstly, the standard errors of the coefficients are higher than normal.

The increased probability of type two error, which is failing to reject a false null hypothesis, is a result of multicollinearity. This is the most significant consequence, as it prevents the Ordinary Least Squares method of estimation from running. Consequently, an accurate regression is not possible. To demonstrate multicollinearity among independent variables, a correlation coefficient matrix is utilized.

The presence of multicollinearity is indicated when the absolute values exceed those in the correlation matrix. Table

3 shows the correlation between each independent variable. A t-test revealed that FUN, MIL, SPD, SBT, ROD, DRIS, SUV, and DPM coefficients were not significant at a 5% level. Hence, it cannot be supported that any of these variables have a significant effect on the dependent variable FCA. The initial coefficient to fail the t-test was FUN. At a significance level of 5%, funding per mile of highways in 2003 (measured in thousands of dollars) divided by total road length miles (in thousands) for each state does not significantly impact FCA. The second coefficient to fail the t-test was MIL.

At the 5% level of significance, the total average vehicle miles traveled per registered motor vehicle in each of the fifty states (in thousands) in 2003 does not significantly affect the dependent variable FCA. The third coefficient of the variable that failed the t-test was SPD. Additionally, at the 5% level of significance, the urban interstate speed limit in each of the fifty states (in miles per hour) in 2003 does not significantly impact the dependent variable FCA. The fourth coefficient of the variable that failed the t-test was SBT. Furthermore, at the 5% level of significance, the seat belt fine amount for each of the fifty states (in dollars) in 2003 does not have a significant influence on the dependent variable FCA.

The t-test did not reveal any statistical significance for the fifth coefficient of the ROD variable. Using a 5% significance level, the proportion of roads in very good and good conditions in 2003 (obtained by dividing the total amount of very good and good roads by the total number of roads in each state)

does not have a significant effect on the dependent variable FCA. Similarly, there was no statistical significance found by the t-test for the sixth coefficient of the DRIS variable. Considering a 5% significance level, the percentage of drivers over age 65 in each state in 2003 does not significantly impact the dependent variable FCA.

The t-test conducted at a 5% significance level revealed that the seventh coefficient (percentage of sport utility vehicle ownership in each state in 2003) and eighth coefficient (number of licensed drivers per square mile) were not statistically significant. Hence, these variables do not significantly affect the dependent variable FCA. Out of the ten independent variables and their coefficients, only three displayed statistical significance according to the t-test.

The SAF variable's first coefficient represents the representation of funding programs per registered motor vehicle in 2003 for the Federal Highway Safety Program. The calculation involves dividing the total amount of federal funds allocated for safety programs (in thousands of dollars) by the total motor vehicle registrations (in thousands of registered drivers) in each state.
According to regression analysis, a thousand dollar increase in funds for the Federal Highway Safety Program results in a significant decrease of 14.67 deaths per 100,000 registered vehicles in fatal car accidents.
The second coefficient pertains to the GAS variable, which indicates the average price of unleaded fuel (in dollars) in each state during 2003. This variable also holds significance at a level of 5%.

According to the regression analysis, an increase of one dollar in gas prices leads to a reduction of 3800.26 deaths in fatal car accidents per 100,000 registered vehicles. The variable DRIY passed the t-test with a coefficient and significance

at the 5% level, indicating that the percentage of drivers under 25 in each state has a significant impact. Additionally, the regression analysis reveals that for every one percent increase in drivers under 25, there is a decrease of 136 fatal car accidents per 100,000 registered vehicles.

The coefficient DRIY in equation (1) has an unexpected sign, leading to 85 deaths. This discrepancy is attributed to omitted variable bias, which occurs when an important variable is excluded from the model. Consequently, the adjusted R2 is low. Furthermore, the low correlation observed in the multicollinearity section of this paper is also likely a result of omitted variable bias.

After discovering these results, I included the variable DPM into equation (1) in order to compensate for the low adjusted R2. Introducing this new variable had a minor positive effect on the regression results. VI. Conclusions The data findings are intriguing, as the analysis indicates that the model had no issues with heteroskedasticity or multicollinearity.

It is uncommon for a model to not have heteroskedasticity or multicollinearity. While estimating the model, it was anticipated that some independent variables' coefficients would show high correlation. Except for FUN and DPM, all independent variables' coefficients had a correlation of less than |.70| with each other. Surprisingly, out of ten independent variables, only three coefficients were found to be significant in explaining fatal car accidents' determinants.

The significant coefficients of the independent variables SAF, GAS, and DRIY in my model indicate unexpected findings. Contrary to what the media portrays about hazardous young drivers, the actual sign of DRIY is negative. According to my model, an increase in the number of young drivers leads to a decrease

in fatalities. This claim can potentially be supported by empirical literature, although there may be disagreements.

The coefficients of the independent variables SAF and GAS are significant in this model. Contrary to DRIY, increasing funding for highway safety programs and gas prices leads to a decrease in fatal car accidents, confirming my hypothesis. It is important to note that my model only captures a portion of the data determining fatal car accidents. Some data, such as the percentage of day time versus night time driving for drivers in each state, could not be processed due to its elusive nature.

Factors like this contribute to the understanding of what causes fatal car accidents. The objective of my investigation is to uncover the determinants of fatal car accidents. Though I might not have covered all variables and their coefficients, I now possess more knowledge about fatal car accidents and their causes. Incorporating econometrics and literature has provided insight into the reasons behind certain fatal accidents and strategies for prevention.