(Craig Linder / States News)The federal government's effort to give rollover ratings to automobiles does not give consumers enough information about vehicle safety and should be revised, a new report claims.A National Research Council study releasedFeb. 21 also says that the National Highway Traffic Safety Administration (NHTSA) should begin using actual roadway tests to determine the likelihood that a vehicle will be involved in a rollover crash.NHTSA has used a five-star system to rate vehicles' tendency to roll over since 2000. In the 2001 model year, most sport utility vehicles received two- or three-star ratings while most passenger cars, which have a lower center of gravity, received four or five stars, the highest ratings.But having only five levels in the rankings means that vehicles with different rollover tendencies can receive the same number ofstars, according to the report. Instead, the report said, NHTSA should move to a system with more rating categories or a numerical score that more clearly distinguishes among vehicles' rollover risks.NHTSA defends its rankings, saying it gives consumers enough information to make a decision. "We did an extensive amount of research before we made the decision to go with five stars," NHTSA spokesman Rae Tyson told Craig Linder of States News. "It's something everybody's familiar with. Everything from crash test scores to restaurant reviews to moviereviews use five-star ratings."But it is the five-star rating system's very simplicity, that irks automakers. "You can't rate a vehicle's propensity to roll over the same way you rate movies," said Eron Shosteck, aspokesman for the Alliance of Automobile Manufacturers. "The star rating system doesn't provide consumers with the kind of useful information they may be looking for."