Skip to main content

What Crash Test Scores Mean

(updated May 5th, 2009)

Update: In October 2010, the National Highway Traffic Safety Administration revised its Five-Star Safety Ratings. Click here for the most current information.

It's one of everyone's worst nightmares: the phone call where you learn that a loved one has been in an accident. I know. It happened to me.

It was a hot summer day and I arrived home around 5 p.m. to discover the answering machine light blinking. I pushed the button to hear my husband Jeff's voice telling me he had been in a car accident and I needed to come give him a ride home. His voice was not that of the usual, confident man I knew. It was tentative and shaky. And confused. He was unsure of where he was, even though it had been his daily route to and from work for years. The staccato phrases were punctuated by long pauses until he got out his location, the fact that he was waiting for a tow truck and needed to talk to the police who had responded to the scene. Then, finally, were the words I was waiting to hear, "I'm OK." Unfortunately, he didn't sound OK and I was worried.

See Edmunds pricing data

Has Your Car's Value Changed?

Used car values are constantly changing. Edmunds lets you track your vehicle's value over time so you can decide when to sell or trade in.

Price history graph example

In the 40 minutes it took for me to drive to the accident scene, I comforted myself by thinking of the car he was driving, a 2001 Volvo S80. We knew when we purchased it that it had a strong safety record. It had received the top ratings in both the federal government and the insurance industry's frontal crash tests. During the drive, I remembered back to the day Jeff and I looked up the crash test scores on the Internet before we made the final decision to purchase the car. Seeing the five-star government rating, Jeff had said, "Hopefully these stars will prevent me from seeing stars if we ever have an accident."

When I arrived at the accident scene and saw the car — its front end crumbled, windshield cracked and airbags hanging like deflated balloons — I realized what crash test scores really mean: the difference between a minor injury and a serious one that affects you and your loved ones.

Keeping Score

When it comes to crash testing, there are two organizations that keep score: the National Highway Traffic Safety Administration (NHTSA) and the Insurance Institute for Highway Safety (IIHS). Both organizations began looking at automobile safety in the 1960s when the public became more aware of the issue as a result of the book, Unsafe at Any Speed, in which author Ralph Nader exposed the safety issues associated with the Chevrolet Corvair. The first crash tests, however, weren't conducted by NHTSA, which is a division of the Department of Transportation, until 1978. Those tests are referred to internally as the New Car Assessment Program, or NCAP, but are also called the "government 5-star ratings" when mentioned in consumer materials. The IIHS, which is supported by automobile insurance companies, didn't begin its crash testing for consumers until 1995.

Currently, NHTSA and the IIHS do both frontal and side crash testing, although each conducts its tests somewhat differently. NHTSA also has a rollover test, while the IIHS assesses seat and head restraints in rear crashes and bumpers in low-speed crashes. When it comes to selecting vehicles, both organizations choose vehicles to test that are the most popular with buyers and have structural or safety features that have been changed from the prior model year. These vehicles, combined with the results from so-called "carry-over vehicles" (those that are largely unchanged from the prior model year), mean that the two groups provide crash test ratings for the majority of the vehicles for the current model year.

Understanding the Tests

While automakers often promote high crash test scores in marketing materials for certain vehicles, the scores don't mean much unless you understand how they are calculated. If you are using this information to help you with a purchase decision, then understanding this data, whether the results are positive or negative, is especially important. Let's take a look at each of the areas where crash testing is currently done.

Frontal Tests

While both organizations do frontal crash tests, the two tests are quite different. In the NHTSA test, two crash test dummies the size of average adult men are placed in the driver and front-passenger seats secured with the vehicle's seatbelts. The vehicle is then crashed head-on into a fixed barrier at 35 miles per hour, which is the equivalent of two vehicles of similar weight hitting each other head-on at 35 mph. The force of the impact to the dummies is measured, and NHTSA gives the vehicle a star rating based on the percent chance of serious injury (meaning an injury that requires immediate hospitalization and may be life-threatening) to the head and chest. NHTSA's star ratings are as follows:

5 Stars = 10 percent or less chance of injury
4 Stars = 11-20 percent chance of injury
3 Stars = 21-35 percent chance of injury
2 Stars = 36-45 percent chance of injury
1 Star = 46 percent or greater chance of injury

The frontal test that the IIHS conducts is offset; meaning that only one side of the vehicle's front end is hit. The vehicle being tested strikes a deformable barrier on the driver side at 40 mph, which means the forces are similar to a frontal offset crash between two vehicles of the same weight that are each traveling at just under 40 mph. About 40 percent of the front end of the vehicle is impacted.

The IIHS ranks the vehicles it tests in one of four positions: Good, Acceptable, Marginal or Poor. Ratings do not correlate to a chance of injury as in the government's test, because the IIHS is assessing more than just occupant injury. It is also looking at how well the vehicle structure performs and the movement of the dummy, such as a partial ejection from the vehicle.

While it would seem at first glance that the two tests are measuring basically the same issues, the fact that the tests are conducted very differently means that the results tend to be quite different. Vehicles are designed today so that the energy that is created in a collision is absorbed through the body of the vehicle and around the occupants. This is done quite successfully in most vehicles today when a vehicle is hit head-on, but what this means is that the force put on the restraints to hold the passengers in place is much greater. "The government test emphasizes how well the restraint systems manage the energy of the occupant," explains Adrian Lund, chief operating officer of the IIHS. "Our frontal offset crash test means that part of the vehicle is managing all the energy, and we know from real-world data that intrusion (of the vehicle into the passenger compartment) is the key in this type of collision."

As a result, both organizations feel its respective frontal tests are complements to each other, not competitive. Nathaniel Beuse, division chief of NHTSA's New Car Assessment Program says that consumers should use the results from both tests "together to assess overall frontal crash test safety in terms of the effectiveness of restraint systems and the integrity of the occupant compartment."

It's important to note that both tests can only be used to get an idea of how the vehicle would perform in a collision with a vehicle of similar size and weight or in a single-vehicle collision, which results in essentially the same forces as a collision with a similarly sized vehicle. They cannot be used to assess how a vehicle would fare if it collides with a vehicle that is significantly different in size. "Most people think that the majority of crashes are with other vehicles, but a lot of the most severe crashes are single-vehicle crashes, so the results are more relevant than you might think," says the IIHS's Lund. According to NHTSA's statistics, about half of all occupant fatalities in 2003 (the most recent statistics available at press time) were from single-vehicle collisions.

Side-Impact Tests

As with the frontal tests, the side-impact tests that the two groups conduct are also quite different. Both tests simulate the type of side collision that would typically occur in an intersection, by crashing a deformable barrier into the vehicle being tested. In the NHTSA test, two dummies that represent average-sized men are placed in the driver seat and in the rear, directly behind the driver. A 3,015-pound barrier is then slammed into the vehicle at 38.5 mph. The force of the impact to the dummies' head, neck, chest and pelvis is measured, but star ratings indicate only the chance of serious injury to the chest. Head injuries, which are not included in the star rating, are reported separately as what NHTSA calls a "safety concern" if the likelihood of head injuries is considered excessive. "Since all tested vehicles are impacted by the same-size barrier, it is possible to compare all vehicles with each other when looking at side crash protection ratings," says Beuse. NHTSA's star ratings are as follows:

5 Stars = 5 percent or less chance of injury
4 Stars = 6-10 percent chance of injury
3 Stars = 11-20 percent chance of injury
2 Stars = 21-25 percent chance of injury
1 Star = 26 percent or greater chance of injury

The IIHS test differs from the government test in the type of barrier that is used, the size of the dummies placed in the vehicle and what the test measures. Using a "Good" to "Poor" rating system, the group measures the potential of injury to the head, neck, chest, abdomen, pelvis and femur, and gives a rating based on the performance in all of these areas. The IIHS uses two dummies that represent small women or 12-year-old children (5 feet tall and 110 pounds) and places them in the driver seat and in the rear seat behind the driver. "The occupants that are most likely to be struck in a side-impact crash tend to be small-statured people, so we wanted our test to assess the injury to those types of people," says the IIHS's Lund.

In addition, the IIHS's barrier has a different shape and weighs more. The group uses a 3,300-pound deformable barrier that is taller and is shaped like the front of a pickup or SUV that is propelled into the side of the test vehicle at 31 mph. "The differences in our barrier mean that intrusion (of the barrier into the side of the test vehicle) is larger and more severe than in the government test," says Lund. The Institute's test is so severe, in fact, that it is unlikely people who experienced such a crash in the real world would come away free of injuries. The group looks for side-impact protection that allows the occupants to survive these types of crashes without serious injury.

Rollover Tests

Currently, NHTSA is the only group conducting rollover tests. The federal agency originally began assigning vehicles a rollover rating based on a mathematical calculation that looked at weight, width and center of gravity to create a statistical likelihood of a vehicle rolling over. The measurement, which NHTSA calls the Static Stability Factor, was criticized widely because it did not simulate any real-world driving situations that would cause a vehicle to roll over.

In 2004, NHTSA began using a dynamic test for rollovers and the star ratings it assigns are now the result of a combination of both the static and dynamic tests (though greater weight is still given to the static test). The dynamic test is done with a weighted vehicle that simulates a load of five passengers and a full tank of gas. The vehicle is driven to simulate an emergency lane change and instruments measure the tire movement. If two of the tires lift at least 2 inches off the pavement simultaneously, then it is considered to have "tipped up," the precursor to rolling over.

Some 2004 model-year vehicles and all 2005 model-year vehicles and beyond are rated using the results of both the static and dynamic measurements. NHTSA's star ratings for rollovers are:

5 Stars = 10 percent or less risk of rollover
4 Stars = 10-20 percent risk of rollover
3 Stars = 20-30 percent risk of rollover
2 Stars = 30-40 percent risk of rollover
1 Star = 40 percent or greater risk of rollover

Rear Crash Tests

Only the IIHS assesses rear crash protection currently. NHTSA does not study these types of crashes because they represent only a small risk of death — only about 5 percent of belted occupants died in rear-impact automobile collisions in the United States in 2003. "We focus on those collisions which represent the greatest proportion of serious and fatal injuries," explained NHTSA's Beuse.

While deaths from rear-end crashes are relatively low, rear crashes are one of the most common causes for neck injuries commonly known as whiplash and, as a result, the IIHS feels it's important to study them. "We want head restraints to be able to protect people of all sizes," says Lund.

As in the government's rollover test, the IIHS uses both static measurements and a dynamic evaluation to rate how well a car's seat and head restraints support a dummy in a simulated rear-end crash. To begin, researchers place a dummy the size of an average adult male in the seat with the seat back adjusted at about 25 degrees — a typical seat back angle. They look to see if the head restraint is 3.5 inches or less from the top of the head and less than 4 inches from the back of the head. If the head restraints are adjustable, then measurements in the down position as well as in the most favorable position are taken. As in other IIHS tests, a rating ranging from "Good" to "Poor" is given.

Restraints that receive a "Good" or "Acceptable" score move on to the dynamic test. In this test, a dummy the size of an average adult male is placed in the vehicle's seat, which is then placed on a sled that simulates a stationary vehicle being struck from the rear at 20 mph by a vehicle of similar weight. Instruments measure the impact on the head, neck, spine and torso, and ratings are assigned based on the results. The static geometric measurements and the dynamic test results are combined for an overall rating.

Low-Speed Bumper Tests

The IIHS also conducts one other type of crash test that the government does not: low-speed bumper tests. While these tests don't really assess safety, they can be of interest to consumers in that ratings are assigned based on the costs to repair the vehicle. In this test, the vehicle is crashed four separate times at 5 mph — both front and rear bumpers into a flat barrier, the front bumper into an angle barrier and the rear bumper into a short pole. Ratings are then given on the usual "Good" to "Poor" scale based on repair costs, which often come in at over $1,000.

To make this test even more applicable to the real-world situations faced by many drivers, the Institute recently conducted a series of tests of 10-mph crashes between the front bumper of a car and the rear bumper of an SUV using the same brand of vehicles and then vice versa. "We paired vehicles from a single manufacturer because we thought that, at a minimum, automakers should be paying attention to the compatibility of the bumpers across their own fleets," says Lund. The Institute learned that the mismatch in bumper heights made repairs even more costly, especially for the owners of the passenger cars. For example, when a Dodge Stratus hit the rear end of a Jeep Grand Cherokee in the test, the result was $3,256 in damage for the car and $1,279 in damage for the SUV. What's more, some of the crashes resulted in leaks to the radiators that would have prevented the owners from driving away.

Using Test Scores Before You Buy

While you may not place safety over performance or comfort and convenience features when buying a new car, taking a look at the crash test scores — all of them — can certainly be an eye-opener when you are narrowing down your choices. Because each of the tests is conducted quite differently, each test can be considered as a separate measure of the safety aspects of the vehicle. For example, a high rating in the government's frontal test and a high rating in the IIHS's frontal offset test mean two very different things and should be judged accordingly. After all, when the safety of your loved ones is at stake, you don't want them seeing stars if they are in an accident.

To review crash test ratings for specific vehicles, visit NHTSA's site at or the IIHS at