Holiday Traffic Fatality Estimate

Independence Day Holiday Period Estimate for 2023

A frequently asked question is “How much more dangerous is travel over the Independence Day holiday?” Two aspects of this question must be considered: “Compared to what?” and “What about changes in the amount of driving?”

NSC compares the holiday to periods of similar length, specifically from 6 p.m. Friday to 11:59 p.m. Tuesday, during the weeks immediately before and after the Independence Day holiday. This chart shows the fatality data from FARS for 1995 to 2021 for comparable holidays. The average number of Independence Day traffic deaths over the last six 4.25-day holiday periods is 16.2% higher than the average number of traffic deaths during the comparison periods (577 vs. 497 deaths). The difference between these two means is not statistically significant at the .05 level.

Source: NSC analysis of NHTSA FARS data

NSC also compares the Independence Day holiday to other holiday periods. When comparing holiday periods of different lengths, an average fatality per day rate is used. The Independence Day holiday, along with the other summer holidays (Memorial Day and Labor Day) tend to have the highest average fatality rates per day. Thanksgiving, Christmas, and New Year’s Day holidays tend to have lower average fatality rates per day.

Source: NSC analysis of NHTSA FARS data

The second question concerns changes in the amount of travel or exposure. NSC is not aware of any data system that tracks changes in vehicle miles of travel by day of the year on a national basis. Lacking an objective measure of exposure change, NSC assumes travel is greater on holidays than on equivalent non-holidays.

If the assumed travel increase exceeds 16.2%, then the risk of dying in a traffic crash during the Independence Day holiday period is less than during comparable non-holiday periods. If the travel increase is less than 16.2% or if travel is actually lower, the risk of dying on the holiday is greater than during comparable periods.

Estimate methods

The objective is to estimate the number of deaths that will occur in traffic crashes during the Independence Day holiday period based on data available several weeks before the holiday. The estimate developed by NSC includes all traffic deaths from crashes during the holiday period.

The general procedure involves three steps:

• Historical data are used to determine the average fraction that holiday fatalities are of the total motor-vehicle deaths for the month
• Total traffic deaths for the coming month in which the holiday falls are estimated using a time series forecasting model
• The projected total for the month is multiplied by the fraction to obtain the holiday estimate

Holiday as percent of monthly fatalities: Total July motor-vehicle deaths are calculated using the latest four years of final data available from the National Center for Health Statistics. Traffic fatality estimates for the Independence Day holiday period are calculated using data from NHTSA FARS.

The table shows total motor-vehicle fatalities for July and traffic fatalities from crashes that occurred during the holiday period. For the previous six 4.25-day holiday periods, fatalities from crashes during the Independence Day holiday period averaged 15.3% of the total fatalities in July.

A time series model was developed to forecast an estimate of total traffic deaths for July 2023. An exponential smoothing model with seasonality was constructed based on monthly traffic deaths recorded through April 2023. This model was chosen because of the seasonal pattern in traffic deaths. The model forecasts total traffic fatalities for July 2023 to be 4,057. Multiplying the projected total fatalities by the fraction obtained in the first step gives an estimate of 619 traffic fatalities from crashes during the holiday period.

The 90% confidence interval for the estimate of total July motor-vehicle deaths is 3,600 to 4,514. If we assume the fraction of July deaths that occur during the Independence Day holiday period is normally distributed, then the 90% confidence interval for that fraction is 14.3% to 16.2%. Combining these two estimates gives the confidence interval for the Independence Day holiday period estimate: 515 to 731 traffic deaths.