Scott Burgess

Not all auto defects are created equally

Numbers are a beautiful thing and, for some, life itself. Pythagoras, the slightly insane Greek philosopher who dreamed about triangles, believed that life was numbers, literally.

In the digital age, carmakers love numbers nearly as much as Pythagoras. Last week, J.D. Power and Associates released a slew of new numbers in the form of its Initial Quality Study, outlining all of the problems (both mechanical and design related) consumers found in their cars last year.

For the consumers, the news is extremely good. More than 80,000 respondents with new cars helped provide proof that Detroit's carmakers build at the same quality level as everyone else. The survey also revealed that the gap between the No. 1 brand and the 37th has shrunk to historic levels. Everyone, it seems, has figured out how to produce well-made vehicles.

But just like a² + b² = c², there's a little craziness behind the figures.

While the Initial Quality Study serves as the messenger, the hair-splitting assumptions we take from this survey are sometimes suspect. The higher a particular brand ranks on J.D. Power's study, the higher its assumed quality: Lexus (which landed at the top of the list) has fewer problems than Mini, which landed at the bottom.

That's why so many news outlets, including The News, list the brands according to how they ranked and comment about how much one brand moved up or another one dropped. J.D. Power moves metal because it has become synonymous with quality.

This should move American consumers, who have never been the greatest judge of quality, to buy the "best" over the "worst." But as everyone's quality improves, the J.D. Power study instead becomes a mirror of the brands reflecting the people who buy them.

Indeed, to paraphrase Pythagoras: "This just doesn't add up."

While there are true values to the Initial Quality Study, its scoring system feels flawed: Reality and perception have the same value in the survey.

For example, if a new car owner cannot correctly operate the stereo on his Audi A4, the complaint becomes "the stereo is not designed very well." That goes down as a mark on Audi's permanent record.

Now if another consumer complains that his transmission fell out 60 days after buying his new SUV because it was improperly bolted to the engine, that is one point against that carmaker. Each problem is worth "one."

J.D. Power says a consumer complaint about a design defect does carry as much weight for the purposes of this survey. A design cannot change and thus the consumer is stuck with the problem -- even if it's only perceived or, worse, self-inflicted -- such as never reading the owner's manual, which, carmakers say, no one does.

Carmakers must attempt to change that "defect" by making everything so simple that people who shouldn't hold licenses can figure it out.

Furthermore, the scoring system is written in the form of the number of problems per 100 vehicles -- which statistically works well as long as the sample is large enough to make it relevant. If we were to do the math, this means if you had 100 Lexus cars and trucks in a parking lot, with 100 owners, 84 would offer a single complaint, while the remaining 16 would stand around and exchange Zagat's recommendations.

Covering the bases

But complainers are like alcoholics: They never stop at just one. J.D. Power covers this base by increasing the sample size and hoping everything averages out. But the parking lot of Lexus owners could very well have seven owners complaining 12 times each and the other 93 saying nothing. The score would be the same, but the result feels much different.

Allow me play with the decimal point. Since people don't buy 100 vehicles, what are the problems per vehicle -- and I know my math is fuzzy here. Using the J.D. Power figures, Ford averages 1.02 problems per vehicle. Toyota averages 1.01 problems per vehicle.

Even J.D. Power officials will tell you, the difference of 1 in its scoring system is statistically meaningless, meaning Ford and Toyota statistically tied when comes to initial quality.

Chevrolet and Suzuki, each with 1.03 problems per vehicle, are only .01 higher than Ford, making them statistically the same as the Blue Oval. So does that mean Chevy is on par with Toyota if it is statistically the same as Ford? And if Toyota is tied with Mercedes-Benz, does that in turn mean Suzuki builds vehicles with the same initial quality as Mercedes, when the difference is .02 problems per vehicle? Certainly no Mercedes owner wants to admit that car is nearly the same quality as a $15,000 Suzuki.

Furthermore, since all consumer complaints carry equal weight and the best and worst get closer and closer, it's more difficult to interpret the data.

That's why a shift in the criteria can have a large effect. A few years ago, J.D. Power began to separate design defects and equipment defects on its durability study. BMW moved from the bottom of the list to the top.

Then there are my questions about the psychology behind the answers. Often they will break down the demographics and psychographics of people buying their cars and trucks.

Does a vehicle's design and style appeal to different groups more prone to complain?

Would a Lexus owner ever admit something was wrong with his vehicle, since it has a longtime record of topping the J.D. Power survey?

Are the expectations so low for Suzuki customers that they're just happy to have something that starts?

Is a diehard BMW enthusiast perhaps more critical of his 335i because he loves it so much?

Do people sometimes find something wrong with a vehicle because they feel obligated to at least look like they're answering all 200-plus questions? After all, no vehicle can be perfect, or there wouldn't be a need for the survey.

In fact, J.D. Power has provided an invaluable service to carmakers and consumers for 23 years -- improving the initial quality of every vehicle.

The annual survey answers innumerable questions for carmakers and helped them all better address consumers' wants, expectations and needs.

The Initial Quality Study empirically shows that Detroit's quality is on par with any carmaker out there. And it is -- but I knew that before the J.D. Power released its numbers.

Quality and style are not the same thing, and eventually someone will figure out how to quantify the latter. Quality may get people to the dealership, but style is what will get them to sign on the dotted line.

And while surveys help carmakers, the only figure that really matters is the bottom line, not just its quality score.

Now, monitor quality

All carmakers will continue to strive to build the best vehicles possible. In many ways, the quality question has been addressed, though it needs to be monitored by groups such as J.D. Power and Associates.

The survey should and needs to remain part of the American car buying experience. But it needs to be placed in the proper context.

I drive hundreds of new vehicles every year and if questioned long enough, I could find two things wrong with every one of them -- meaning they would score worse than the 1.65 problems each Mini collected.

Numbers don't lie, but how we use them is up to us.

sburgess@detnews.com (313) 223-3217