This morning I received an email from Britain’s Auto Express magazine, announcing that the results to “Britain’s top car reliability survey” had been posted here.

Intrigued, I clicked through to check it out, and found that some European models do quite well. At the model level, the top four are the Skoda Octavia (related to the VW Jetta), Land Rover Discovery (LR3 in the US), Mercedes E-Class, and MG ZT (not sold here). The Lexus IS manages fifth. At the manufacturer leavel, Skoda (part of VW), Jaguar, and BMW rank second, fourth, and fifth, respectively.

Much different results than I’ve seen in any US vehicle reliability survey.

Once again, it pays to look at what was measured by the survey. I couldn’t find the actual questions, but “results by category” are posted. Categories include, in addition to reliability: performance, build quality, running costs, comfort, practicality, ease of driving, braking, handling, and ride quality.

When a survey includes ten categories, only one of which is reliability, how is it still a “reliability survey?”

If you look at reliability alone, 18 of the top 20 models are Asian, the Skoda Octavia ranks 20th, and the Land Rover Discovery (LR3) ranks 39th.

On top of this little sleight of hand, I suspect that the survey measures reliability by asking something along the lines of, “How reliable is your car?” Such wording makes the results highly subjective. It’s measuring owner opinions, not actual repair rates.

The results are still quite interesting, just realize what they actually measure.