JD Power Dependability Study Continues To Rank Minor Tech Issues The Same As Total Vehicle Failures - Should They?

I've done a JD Powers survey when I got my latest vehicle in May 2015. They simply have a long list of categories and questions. They call these "complaints". Find the downshifts in your automatic transmission a little rough? That's a complaint. Find the computer infotainment system confusing or requiring too many steps? Find the climate control system difficult to use? That's a complaint. Anything the new owner doesn't like is a complaint.

Then there are defects, such as leaking seals and water getting into the car. The survey views those as complaints too. This information is not only compiled by JD Powers, they also send it to the manufacturer.

I believe it's best to keep these things as one combined number because it would be easy for a manufacturer to hide behind the principle of "nothing's broken, so the car is perfect", and in doing so, ignore the fact that owners are unhappy with some aspects of their vehicles. It's great that Audi gets lots of feedback from owners that they don't like their infotainment system (a common and frequent complaint). It's great that lots of people told BMW they didn't like the self-cancel-only turn signals (no manual cancellation available).

For the people who consider the JD Powers data in their purchase decision, it's good to know that some brands, and some models have more satisfied customers (fewer complaints) than others.

They absolutely need to have weighting between some categories. An issue with Bluetooth connectivity should not have the same weight in the results as an engine or transmission issue. The current statistics have no weighting whatsoever, so someone that doesn't understand technology might complain mightily about something that really has nothing to do with Dependability.

Most American are too technologically illiterate to operate the simplest of tech features in their cars, so my opinion is that the majority of complaints on car technology features should not be counted the same as a mechanical issue.

For all we know, those with less pp100 might have more expensive and severe issues reported than a brand at the industry avg.
How can anyone know what the true reliability is for each brand if CR doesn't put more weighting to their study.
This can be a very complicated report if CR (or anyone for that matter) wanted to be detailed.
My 2 cents
mechanical breakdowns: weight = 10 (i.e. transmission fails)
mechanical annoyances: weight = 2 (i.e. transmission changes are rough)
understanding the tech: weight = 1/each feature not understood

The last bit can have a curve where once you submit your 8th "feature not understood" the system should decide you're an imbecile and doesn't count your report at all.

This isn't in reference to CR, although the same thoughts apply. This is JD Powers, which is supposed to be able to perform better analysis than this.

JD Power's data analysis is a classic example of flawed study design.

This could be a classic example of using inaccurate terminology. The report describes J.D. Powers' tabulation of customers' *complaints* (or issues), of any sort, and yet the sentence "...Problems with technology continue to affect vehicle reliability..." appears. Reliability might not have anything to do with usability or customer preferences. Does it work as designed, and for the specified period? If so, then it's reliable, even if you don't like it (e.g., climate control system that takes "too many steps" to perform a function, or which requires the user to read the manual to figure it out).

On the other hand, a voice recognition system that misunderstands commands, or a Bluetooth system that doesn't pair with a specific Bluetooth phone can be considered a reliability problem providing that the user manual omits the disclaimer such as "might not be compatible with all phones".

I'm all for reporting all user complaints, regardless of their nature. I also like the idea of knowing the actual defect numbers, where something breaks or does not work reliably during the specified period. Both types of information are useful.