Every year, Northern Virginians bring about 1.5 million cars, trucks and other vehicles to their local service station for a safety inspection required by the state. The annual ritual, marked by long lines and the threat of mandatory repairs, is supposed to be the same throughout Virginia.

Yet compared with the rest of the state, vehicle owners in Northern Virginia are much more likely to flunk the tests, according to a computer analysis by The Washington Post and washingtonpost.com.

Stations in suburban Washington failed vehicles about 40 percent more often than their counterparts elsewhere in the state, and more than twice as often as stations in Southwest Virginia, the analysis found.

At the same time, the study found wide variations in failure rates within the region itself: Nearly 20 stations rejected at least one of every five vehicles inspected, while scores of others passed all, or nearly all, the vehicles they examined.

The disparities, the system's critics say, reflect the likelihood that some stations are failing cars just to charge for unnecessary repairs, while others are rolling cars out without careful examination just to collect the $10 fees.

"It certainly leads you to believe something is amiss here," said Mark Edwards, managing director of traffic safety for the American Automobile Association.

But unlike some other states with inspection programs, Virginia does not keep the records that would allow detailed audits and does not use anonymous test cars in a systematic way to check stations' honesty.

Virginia State Police Capt. Steve Flaherty, head of the vehicle safety program, said that although he has confidence in the system, "I'm not so naive to believe there are not some problems out there. I'm sure there are stations out there that are not treating the public the way we want them treated."

Every year, the owners of the more than 6.5 million vehicles registered in Virginia must have their cars, trucks or vans checked at a certified safety inspection station. The stated purpose of the program is to improve highway safety by reducing the chance of mechanical failure. If a vehicle passes, it gets the required approval sticker on the windshield; if not, it gets a failure sticker, and the owner must fix it within 15 days.

(Vehicles in the District submit to combined safety and emissions inspections every two years by the municipal government; Maryland requires safety inspections by privately run stations when a vehicle's title is transferred or when a vehicle's owner moves into the state.)

Virginia State Police assign about 50 troopers full time to supervise the 4,300 certified gas stations, repair shops and automobile dealers. Troopers are expected to visit every 45 days to examine paperwork and equipment. Police sometimes send out test cars but nearly always to stations already suspected of having a problem.

About 1 percent of inspectors or stations are suspended during a given year, usually for matters unrelated to rejection rates. Last year, state police received about 6,000 complaints about safety inspections, Flaherty said. About 5,000 were generated by the state because of improperly filed paperwork and other administrative problems; about 1,000 were complaints from the public.

To state officials, the tiny proportion of complaints and suspensions indicates that the system works well. But others say that many customers don't know where to complain. For many drivers, the annual ritual is marked by uncertainty--a feeling that it's all just a bit of a crap shoot.

"The whole thing is sort of suspicious," said Veronica Bennett, of Clifton, as she awaited inspection results on her 1994 Plymouth Voyager at a Manassas service station. "I think there's many instances of stations flunking more people than they should. Others just kind of wave you through. It's hard to tell whether they're being fair or not."

Most complaints are settled in favor of the station or inspector. In 1998, 112 inspectors and 74 stations were suspended. Flaherty estimated that fewer than 10 inspectors were permanently kicked out--all of them for selling or giving away approval stickers.

L.R. "Ray" Trail, a trooper in the safety division's Fairfax District, said the state's position is that stations rarely do anything to endanger their livelihood. "If there's a benefit of the doubt, we give it to the inspector," Trail said. "It does bring in work for them, and they don't want to take a chance at losing that license."

Some stations have had more than one complaint filed against them.

At Triangle Amoco on Route 1 in Prince William County, owner Munawar Karim was issued a reprimand in June 1997 for not noting brakes nearly worn down to the metal on a Lincoln Town Car. Three months later, his station was suspended for 45 days when a trooper confirmed that Karim had wrongly flunked one vehicle for headlights and neglected to properly tighten a wheel on another.

Finally, in May 1998, customer Michael Welch complained to police that Karim wrongly flunked one car for bad brakes and another for an exhaust leak without even bothering to lift the vehicles off the ground.

"The brakes were good," Welch, an automobile mechanic, told police. "Nothing was wrong."

The state trooper investigating the case agreed, recommending that Karim be suspended for six months. But regulators overruled the trooper, and more than a year later, Karim is still working as a certified inspector.

Welch could not be located for comment. Karim did not return repeated telephone calls, but he told investigators that one inspection was conducted properly and the second was just a "look-over." Flaherty said the agency reluctantly dropped the matter because it could not determine whether Karim was supposed to be performing full-scale inspections.

Suspicions

Another contention by critics is that the system has built-in incentives for inspectors to cheat.

Stations that flunk vehicles, for example, may repair the cars and set their own rates for the work. Most stations count on such repairs because the mandatory $10 inspection fee doesn't reimburse them for the time it takes to conduct a thorough analysis.

"You want to make money on the repairs from inspections," said Jim Foster, owner of Foster's Exxon in Herndon, another station that had a high rejection rate in The Post's analysis--one rejection for every four to five stickers. "That's just an honest statement." Foster said he conducts detailed inspections and follows the state's rules closely.

Some have raised concerns about the overall quality of the stations that perform inspections. A recent report by the Center for the Study of Services, a District-based research group that publishes Washington Consumer Checkbook magazine, found that inspection stations in Northern Virginia and suburban Maryland had lower customer satisfaction and higher prices than auto repair shops that don't perform inspections.

"A lot of the less-good auto repair shops get into this kind of business because it's a way to create traffic for themselves," said Robert Krughoff, president of the group. "There's good reason to believe that those same kinds of stations would be flunking people to drum up repairs."

The other way to make money is to rush customers through. If a station racks up enough inspections per hour, it can meet or exceed the rates it charges for repairing cars, say station owners and inspectors.

Most inspectors get a commission on each inspection fee, according to both inspectors and station owners, an arrangement critics say could provide an incentive to speed things up and wave through defective vehicles.

"If they can turn out a car every 10 minutes, they're beating the clock and keeping the line rolling and bringing in 60 bucks an hour," said Paul Sisson, an inspector at Gunston's Texaco in Lorton. Sisson said his station's rejection rate, one of the highest in The Post analysis, was attributable to many older, poorly maintained cars in the area.

Lobbyists for the state's service stations contend that the main problem with the system is that the fee is too low and that not raising it could lead to abuse.

"At $10, it's becoming very difficult for an inspection site to break even or attract qualified inspectors," said Bruce Keeney, executive vice president of the Virginia Gasoline Marketers Council.

Theories

Even if the state wanted to check for widespread problems in the program, it could not do so systematically without changing the way it handles oversight.

Nearly all information about inspection stations--the number and nature of complaints, the number and outcomes of inspections--is now kept in paper files that are purged every 18 months. Only shipment records of pass-fail stickers are kept in electronic form.

Flaherty says his division, which has an annual budget of about $5 million, needs up to $7 million for new computer equipment to keep better tabs on the program.

But even though officials acknowledge they do not have the data to explain the pass-fail rates, they say the variations could be attributable to factors other than misconduct. The differences between regions, and even stations, they suggest, could be linked to geographic and economic factors--although some regional differences seem to defy logic.

For example, rural southwestern Virginia, home to many older cars, had the study's lowest rejection rate, while Northern Virginia had the highest, even with its high proportion of newer and luxury vehicles. The Post's analysis also found that many stations in Northern Virginia that issued high percentages of rejection stickers were located within the same Zip codes as stations with much lower rates.

The top station on the list was Dunivin's and Son in Woodbridge, which issued 6,911 rejections and 10,093 approvals from January 1996 to June 1998--meaning that more than 40 percent of the stickers issued were rejections.

Several neighboring stations, however, had much lower rejection ratios. Rollison's Automotive Services Inc., owned by state Del. John A. "Jack" Rollison III (R-Prince William), was nearly 10 times less likely to issue rejections than Dunivin's.

Owner Jeff Dunivin attributed his station's high rejection rate to business from a neighboring used-car lot, and he said he recently decided to stop offering inspections because he didn't attract enough profitable repairs. "We had a lot of trouble with auction cars, and they're never in good shape," he said.

Owners and inspectors at other stations with the highest rejection rates say they are only following state guidelines, rejecting cars with bad headlights, broken windshield wipers and other serious defects that make them unsafe to drive. They said the real problem with the system is that many stations conduct rapid, cursory inspections--approving vehicles that are defective and even unsafe--in order to make money on the $10 fees.

Worth It?

At one time, federal officials were such proponents of state safety inspection programs that they required states to have them if they wanted federal highway funds.

But that rule--never strictly enforced--was eventually scrapped. And in 1989, a study by the National Highway Traffic Safety Administration found that although inspections clearly kept decrepit vehicles off the road, there was little evidence they resulted in fewer accidents. Later reports came to contradictory conclusions, but the government has not resumed active promotion of the programs.

In Virginia, a state commission in 1994 debated axing inspections after hearing complaints that stations rejected too many vehicles and after considering whether the program actually prevented accidents.

But in the face of heated opposition from service stations and traffic-safety groups such as AAA, the commission declined to seek major changes.

Besides Virginia, 22 states have mandatory safety inspections of passenger vehicles by government or sanctioned inspectors, down from 31 in the mid-1970s, according to the American Association of Motor Vehicle Administrators.

Pennsylvania, which has an inspection system similar to Virginia's, also has stronger oversight, including a regular audit system that uses test cars to randomly sample station performance.

"That helps us maintain a fair degree of consistency across the state," said Terry Liller, director of Pennsylvania's safety program.

Liller said the audits have not found wide disparities in failure rates by region or among individual stations like those found by The Post.

"Those kinds of major differences would concern us a great deal," Liller said, "and we would be compelled to figure out why that is and solve the problem."

Hal Straus, washingtonpost.com database editor, contributed to this report.

BREAKDOWN BY REGION

Rejection rates for vehicle safety inspections vary widely from region to region in Virginia, according to a Washington Post analysis of inspection sticker orders from January 1994 to June 1998. The Fairfax District, which encompasses Northern Virginia, finished at the top of the list.

Division I; Richmond: 12.8%

Division II; Culpeper: 12.8%

Division III; Appomatox: 8.8%

Division IV; Wytheville: 6.7%

Division V; Chesapeake: 14.0%

Division VI; Salem: 8.2%

Division VII; Fairfax: 14.5%

NORTHERN VIRGINIA CAR INSPECTIONS

These sixteen stations rejected at least one of every five vehicles inspected during the 2-1/2 years ending in June 1998, according to a Washington Post analysis of state records. Virginia officials say they would expect stations to flunk no more than 10 to 15 percent of vehicles at initial inspection, although they have not conducted studies to determine what the average actually is.

Station: Dunivin's and Son

Location: Woodbridge

Approvals: 10,093

Rejections: 6,911

Rate*: 40.6%

Station: Village Auto Care

Location: Woodbridge

Approvals: 2,880

Rejections: 1,107

Rate*: 27.8

Station: Chantilly Auto Service

Location: Chantilly

Approvals: 7,092

Rejections: 2,514

Rate*: 26.2

Station: Woodbridge Auto Service

Location: Woodbridge

Approvals: 5,299

Rejections: 1,803

Rate*: 25.4

Station: Route 1 Citgo

Location: Woodbridge

Approvals: 10,463

Rejections: 3,471

Rate*: 24.9

Station: Circle Shell Service Center

Location: Fairfax

Approvals: 10,055

Rejections: 3,264

Rate*: 24.5

Station: Foster's Exxon

Location: Herndon

Approvals: 9,200

Rejections: 2,801

Rate*: 23.3

Station: Gunston Texaco

Location: Lorton

Approvals: 5,596

Rejections: 1,664

Rate*: 22.9

Station: Saratoga Mobil

Location: Springfield

Approvals: 10,030

Rejections: 2,957

Rate*: 22.8

Station: Seminary Plaza Exxon

Location: Alexandria

Approvals: 9,349

Rejections: 2,725

Rate*: 22.6

Station: Scottie's Texaco

Location: Manassas

Approvals: 9,668

Rejections: 2,800

Rate*: 22.5

Station: Herndon Mobil

Location: Herndon

Approvals: 12,327

Rejections: 3,218

Rate*: 20.7

Station: Alexandria Texaco

Location: Alexandria

Approvals: 9,214

Rejections: 2,394

Rate*: 20.6

Station: Lake Ridge Mobil

Location: Woodbridge

Approvals: 6,214

Rejections: 1,607

Rate*: 20.5

Station: Gill's Automotive Service

Location: Sterling

Approvals: 8,882

Rejections: 2,227

Rate*: 20.0

Station: Triangle Amoco

Location: Triangle

Approvals: 4,986

Rejections: 1,229

Rate*: 19.8

*Rate is calculated by dividing the number of rejection stickers issued by the total number of stickers issued.

HOW THE ANALYSIS WAS DONE

Virginia does not keep track of how often the state's certified safety inspection stations pass or fail the vehicles they check. State police officials, who oversee the system, keep most inspection records on paper and are unable to systematically compare rejection rates among stations or regions.

But state officials do record on computer the number of pass and fail stickers that each station orders--numbers that, over time, reflect how likely stations are to find problems with the cars and trucks they inspect.

Using those records, Hal Straus, database editor for The Post's Web site, washingtonpost.com, examined how many stickers of each kind were ordered by 1,800 of the state's busiest inspection stations over a 4 1/2-year period ending in June 1998. The Post also analyzed paper documents--from January 1996 to June 1998--from nearly two dozen stations in Northern Virginia that finished highest on the list.

The rejection rates calculated from the sticker-shipment records underestimate the true rates, because flunked vehicles often are approved later by the same station. Consider a station that rejects five of every 10 vehicles: Assuming the same station eventually passes all of them, it would issue five rejections out of 15 total stickers, or 33 percent, despite a true failure rate of 50 percent.

About 900 of the approximately 4,300 stations statewide requested no rejection stickers at all, while scores of others used just a handful.

This group included stations used by the public as well as 300 so-called private inspection stations, which include utilities, other businesses and government agencies such as the state police that conduct their own inspections of their vehicle fleets. Members of this group are never issued rejection stickers because they are presumed not to need them.

To see how many approval and rejection stickers your inspection station was issued in the last four years, visit The Post Web site at www.washingtonpost.com/metro.