Where is a motorist most likely to die in a railroad crossing crash?

• On a rural, straight, level, blacktopped, dry road in Texas with a 55-mile- per-hour speed limit and crossbucks marking the crossing.

When is a motorist most likely to die in a railroad crossing crash?

• On a Friday or Saturday during daylight hours of the winter months.

What are some common characteristics of this motorist most likely to die in a railroad crossing crash?

• A white male, 25 to 34 years old with no college experience and a blue- collar job who reads "Field and Stream" and "Outdoor Life," watches the family channel on television, listens to country music on the radio, and chews tobacco.

Who would ask and answer these questions?

• This information is from a November 1994 National Highway Traffic Safety Administration (NHTSA) Technical Report (DOT HS 808 196) written by Terry Klein, Tina Morgan, and Adrienne Weiner and entitled "Rail- Highway Crossing Safety: Fatal Crash and Demographic Descriptors." The report was one study resulting from the U.S. Department of Transportation's Rail-Highway Crossing Safety Action Plan, which combined proposals from the Federal Highway Administration (FHWA), Federal Railroad Administration (FRA), Federal Transit Administration (FTA), and NHTSA.

Why would NHTSA ask and answer these questions?

• The purpose of this portion of the safety action plan was to research the demographics of motorists killed in railroad crossing crashes to facilitate design and distribution of properly targeted countermeasures that will help reduce railroad crossing fatalities.

Data in this study came from two sources. First, researchers used figures from NHTSA's Fatal Accident Reporting System (FARS) for 1975 thorough 1992. Specifically, they gathered data on fatal rail crossing crashes; that is, crashes on public roadways involving a motor vehicle and a train where the first harmful event for the motor vehicle was collision with the train and the first harmful event for the train was collision with the motor vehicle. The fatality or fatalities must have occurred within 30 days of the crash. As comparison figures, researchers used fatality records for all intersection crashes and for all crashes. Figures for fatal intersection crashes did not include rail crossing fatalities; figures for all fatalities did incorporate rail crossing fatalities.

The second source of information was Claritas, a commercially available data base that divides the United States population into 3,500 groups by zip code. For each zip code, Claritas identifies 62 cluster groups within 15 social groups organized by how urban or rural they are and their socioeconomic status (occupation, income, home value, etc.). In 1987, FARS began including the drivers' zip codes in crash reports. This made it possible for researchers to merge zip codes of drivers involved in fatal rail crossing crashes with information from the Claritas data base to learn where those drivers lived, their socioeconomic status, what television programs they watched, what they read, and what their consumer habits were.

The study reported results in seven categories: fatal crashes, traffic control devices, roadway characteristics, state statistics, involvement by vehicle type, drivers involved, and Claritas.

Fatal Crashes

From 1975 to 1992, FARS statistics showed a downward trend in the number of fatal rail crossing crashes and resulting fatalities. The chart revealed spikes in 1978 and 1984 and a slight upward trend from 1987 to 1989; however, there were 690 fatal rail crossing crashes in 1975 compared with 388 in 1992 and 851 fatalities resulting from rail crossing crashes in 1975 compared with 489 in 1992.

Charting the percentage of fatal rail crossing crashes by month showed slight increases during the winter months, October through March. This contrasted with figures for intersection crashes and all crashes, which showed increases during the summer months.

Like fatal intersection crashes, fatal rail crossing crashes were most common on Fridays and Saturdays (about 17 percent of the total each day) and less frequent from Sunday through Thursday (about 13 percent a day). Figures for all crashes, on the other hand, showed they were more likely to happen on Friday, Saturday, or Sunday and less frequent from Monday through Thursday.

Fatal rail crossing crashes most often happened between 6:00 a.m. and noon. However, total crash statistics showed most roadway fatalities occurred between 6:00 p.m. and 3:00 a.m., with the fewest between 3:00 a.m. and noon. Intersection fatalities most often happened during rush hour, from 3:00 p.m. to 6:00 p.m., with a steady increase from 3:00 a.m. to 6:00 p.m. and a low rate from 3:00 a.m. to 6:00 a.m.

The percentage of fatal rail crossing crashes that occurred during daylight hours was about the same as the percentage of fatal intersection crashes--more than 50 percent. To the contrary, fatal crashes as a whole were more likely to happen during the hours of darkness.

Traffic Control Devices

FARS reporting categories for traffic control devices changed in 1982. Before 1982, there were seven categories of control device; after 1982, figures were given for 16 types.

Eight percent of fatal rail crossing crashes between 1975 and 1981 happened where there was no traffic control device; this figure dropped to 2 percent from 1982 to 1992. The report's writers theorized this probably resulted from traffic control devices being installed at more rail crossings.

Of the 16 types of traffic control device used at rail crossings, one type controlled traffic at more than 30 percent of the crossings where fatal crashes occurred from 1982 through 1992. That traffic control device was the railroad crossbuck. The crossbuck is an X-shaped, black-and-white sign displaying the words "RAILROAD CROSSING." Such a device is passive--it does not change when a train approaches. The warning is the same no matter how imminent the danger. On the other hand, while 42 percent of fatal rail crossing crashes in 1992 happened at intersections with crossbucks, 51 percent of all rail crossings in the United States were controlled by crossbucks in 1992. These figures do not consider the comparative volume of traffic at different crossings.

Rail crossings controlled by flashing lights and gates also showed high crash rates from 1982 through 1992. These are active control devices that change when a train approaches.

Of the fatal rail crossing crashes between 1982 and 1992, only 95 percent happened at crossings where the traffic control device was working properly. This contrasts with intersection crashes, 99 percent of which happened at intersections with properly functioning traffic control devices.

Roadway Characteristics

The largest percentage of fatal rail crossing crashes happened on roads where the speed limit was 55 miles per hour. The percentage of fatal rail crossing crashes occurring on 55-mile-per-hour roads was similar to the percentage for fatal intersection crashes--31 and 29 percent respectively. An even higher percentage of all fatal crashes happened on roads with 55-mile-per-hour speed limits--45 percent. The next highest percentage of fatal rail crossing crashes occurred on roads with speed limits of 25 to 30 miles per hour--26 percent.

The majority of all fatal crashes and fatal rail crossing crashes happened in rural rather than urban areas--56 percent and 63 percent respectively. Although the report did not specify the number or percentage of railroad crossings located in rural versus urban areas, the authors theorized, This pattern likely parallels the rural-urban distribution of rail crossing locations. On the contrary, only 16 percent of fatal intersection crashes happened in rural areas. It is likely this was a result of the relatively high number of intersections in urban areas compared to rural areas.

Fifty-three percent of fatal rail crossing crashes were on rural roads or urban local roads and streets, while only 18 percent of all fatal crashes and 16 percent of fatal intersection crashes were on rural roads or urban local roads and streets. Again, the difference was probably due mainly to the larger number of railroad crossings on rural roads and urban local roads. Other roadways more often have railroad overpasses and underpasses.

State Statistics

With 7.6 percent of the nation's rail crossings, Texas had 9.3 percent of the fatal rail crossing crashes in 1992--the highest percentage rate of any state. In that year, over 50 percent of the fatal rail crossing crashes happened in eight states: Texas, Ohio, Indiana, Michigan, Louisiana, Illinois, Oklahoma, and California.

Involvement by Vehicle Type

Figures showed that about 60 percent of fatal rail crossing crashes, fatal intersection crashes, and all fatal crashes involved passenger cars. More light vans and trucks were involved in fatal rail crossing crashes than in fatal intersection crashes or in all fatal crashes. Researchers suggested this might be a reflection of the preference for such vehicles in rural areas where more railroad crossings are found than in urban areas.

Vehicles involved in fatal rail crossing crashes were more likely to be severely damaged and to suffer an initial impact from the side than vehicles involved in fatal intersection crashes or fatal crashes in general.

Drivers Involved

The group most likely to be involved in fatal rail crossing crashes was between the ages of 25 and 34 (23 percent), and the second most likely was between 16 and 20 (18 percent). As the authors stated, "The 16-20 year old group is particularly high when considering that this age group includes only five years, while the 25-34 year olds include a ten-year span." These two age groups were also those most involved in fatal intersection crashes and all fatal crashes.

Distribution by gender was similar to distribution for all fatal crashes and fatal intersection crashes--about 77 percent males.

Of drivers involved in fatal rail crossing crashes between 1987 and 1992, 24.4 percent had a blood-alcohol concentration (BAC) over .10--the legal intoxication level in many states. The alcohol-involvement level was about the same for fatal crashes overall (26.1 percent) but lower for fatal intersection crashes (12.8 percent). However, in 45 percent of fatal one-car crashes, drivers showed a BAC over .10--a considerably higher percentage than for fatal rail crossing crashes. For all categories, drivers 21 to 24 years old had the highest rate of alcohol involvement in fatal crashes, and 25 to 34 year olds were second highest.

In other FARS categories, figures showed 95 percent of fatal rail crossing crashes happened on straight roads, 75 percent on blacktopped roads, and 63 percent on level roads. Only 6 percent were on divided highways, and 81 percent happened on dry roads. Reports most often indicated failure to yield, failure to obey, and inattentiveness as driver-related causes in fatal rail crossing crashes.

Claritas

Researchers used driver zip codes to merge FARS information with the Claritas data base. FARS did not collect zip codes until 1987; therefore, these figures only included drivers involved in fatal rail crossing crashes from 1987 through 1992. Using data for United States residents over 15 years of age yielded a Claritas base study population of 201,104,559 people--with 2,678 in the study population of drivers involved in fatal rail crossing crashes and 383,172 in the study population of drivers involved in all fatal crashes (excluding rail crossing crashes).

As mentioned earlier, Claritas data is important for developing informational materials appropriate to the driving population most likely to be involved in fatal rail crossing crashes and distributing those materials through the media that is most likely to reach that driving population.

While only about 18 percent of the United States population is rural, about 42 percent of drivers involved in fatal rail crossing crashes fell into a rural classification. Again, the report did not quantify the distribution of railroad crossings between rural and nonrural areas, but the authors concluded, This overrepresentation is likely due, in large part, to their exposure to rail crossing opportunities as a result of geographic locations of these populations.

Sixty-five percent of drivers involved in fatal rail crossing crashes were in two Claritas social categories: rustic living and heartlands. Both groups live in rural areas. Groups that included only small numbers of drivers involved in fatal rail crashes were from elite suburbs and uptown areas, where few railroad crossings exist.

For drivers involved in fatal rail crossing crashes, researchers computed a weighted average figure in each of six Claritas categories: education, household income, family type, occupation, race/ethnic origin, and household size.

Results showed that 60.9 percent had no college experience. This was compared with 58.5 percent for other fatal-crash-involved drivers and 54.5 percent for the United States population as a whole. The median household income for drivers involved in fatal rail crossing crashes was $27,667, compared with $29,649 for those involved in other fatal crashes and $31,900 for the U.S. population. The differences in family type (married, single, with or without children) were not significant. Blue-collar occupations were reported by 32.8 percent of the fatal-rail- crossing-crash-involved drivers compared to 31.2 percent for other fatal-crash- involved drivers and 26.6 percent for the U.S. population. Probably due to the larger number of rail crossings in predominantly white, rural areas, 85 percent of drivers involved in fatal rail crossing crashes were white compared with 83.3 percent for other fatal crashes and 80.1 percent for the U.S. population. Researchers found no significant difference in household size among the three study groups.

Segregating the data for drivers involved in fatal rail crossing crashes revealed that the top six magazine choices of the group were: Field & Stream (65 percent), Outdoor Life (65 percent), Sports Afield (64 percent), Hunting (56 percent), Country Living (52 percent), and Guns and Ammo (49 percent). The top five media choices were: watching the family channel (56 percent), reading fishing and hunting magazines (52 percent), listening to country radio (51 percent), watching quiz and audience-participation shows on television (39 percent), and watching television wrestling (39 percent). A final correlation involved drivers in fatal rail crossing crashes and lifestyle choices as presented by Claritas. The top five choices for those involved in fatal rail crossing crashes were: using chewing tobacco (62 percent), going hunting (52 percent), eating at ice cream restaurants (42 percent), listening to gospel music (40 percent), and eating at seafood restaurants (39 percent).