FKL

1. 2002 Accord EX V6: Dead AT @ 60K. Dead AT @ 98K. Bad power window regulator on Drivers side @80K, bad alternator @95K
2. 04 Accord EX-L I4 5MT: Bad navigation unit @ 55K (replaced courtesy of Honda), bad power window regulator @60K. Lots and lots of rattles.

The idea that a Honda is trouble free "because CR says so" isn't true for everybody. With that said, I still see the two cars as reliable for the most part. No bad engine sensors, no sorts of suspension issues, and no electronic problems (beyond navigation computer).

ivanz

I had an 03 Honda Pilot until I sold it last year (~95k kms). Nothing ever needed to be fixed except a cosmetic issue (chrome was rubbing off the interior driver door handle).
No rattles or any other issues with it at all.

I8ABMR

Im sad to say this but I have to say that I am feeling that the there are certain issues I have with my GS that do show Lexus slipping . Whats the deal with the creaks in the console, dash rattles, and blemishes in my wood steering wheel . The service accuracy has seemed to take a dive at the last 3 visits I have had. I am sorry to say this but I am saying that Lexus is solid but there are other solid brands. They need to get their **** together. This is exactly what the American car companies will try to rub in the face of Lexus. Can you blame them?? I came to Lexus from Infiniti because I expected to be this infalable car. Not so. There is no excuse for interior rattles and noises

I8ABMR

Infiniti at number 4 and Lexus at 7. I guess the 3000 quality assurance employees they hired 2 years ago are finally starting to pay off. I never thought I would see the day that Infiniti produces a higher quality product than Lexus. With the base brands like Toyota I expect reliabilty because they are much more basic than a Lexus. I guess I will be looking at Infiniti and .... ( I am going to dare to say it).... BMW

mmarshall

That is correct. Although there have been some transmission issues in the past, the average Honda/Toyota product, if serviced properly, lasts a good 200K miles, and usually doesn't need major repairs before 100-150K. Honda and Toyota did not get their good reputations just out of thin air....for the most part, they earned it.

RX300-BV

Remember that reliability from Consumer Reports is not determined by the editorial staff but by the subscribers that respond to CR's annual car survey. The editorial staff do recommend or do not recommend products as well as provide the driving score.

Because CR reliability is provided by the CR subscribers, the data is inherently biased to its reader base, which for years has been very pro-Japanese. That doesn't mean the data isn't accurate, but it does give reason to pause and evaluate CR's data in conjunction with other publications like JD Power or True Delta. If an auto company scores well in multiple surveys (e.g. Lexus), it's probably a better indication of reliability than say one that has fabulous scores in one yet abysmal in another (e.g. Scion).

What CR should do is provide (and perhaps they already do) the exact breakout of their data. So if they say the GS AWD has terrible reliability but the RWD has great reliability, how many vehicles (AWD and RWD) is that based on? The other flaw with CR is there is no data validation. A survey taker can put in information on multiple cars, even on a car that person doesn't actually own.

My point really is CR's data is flawed, but no more flawed than anyone else's data, so take it with a grain of salt. It doesn't hurt to compare it with other reliability surveys.

bitkahuna

iTrader: (20)

post of the day. but i think CR's method is MORE flawed than other surveys though because, as you say, the data is constrained to a SUBSET of their subscribers, which for many less popular or production limited vehicles means they're working from MINUSCULE sample sizes. how many CR subscribers have porsche cayenne turbos? 5? 10? how many of those actually do the survey? 2? as you say, they don't publish sample sizes so we'll never know, but when it's constrained to a subscription base, the numbers cannot be large.

these CR survey results appeal to those who end up ranked highly , those who spend money to subscribe to CR, and those who consider themselves 'sensible' and would subscribe to CR, but for whatever reason, don't.

in other words, i think CR's survey is not worth a great deal.

but i do agree with their one finding... chryslers are junk. but bmw's aren't that bad.

The G Man

CR is suppose to have a minimum of 100 data sample before they consider the data valid. For the 2009 survey, I believe 1.3 million CR subscriber respond to their survey, yet some sample size were in the hundreds and some are in the tens of thousands. My problem with this type of method is that the weight of each survey is not equal, meaning one person who have a negative option of a low sample model car can change the final result much more than a person who is evaluating a car with a massive sample weight. This year CR actually broke their own rules by using sample rate below 100 for 3 of Ford's model.

I thing CR does a much better job when they rate things like appliance and tires where they actually test the product in their lab, but obviously, they cannot buy a million and test it. Never the less, they should really improve the way they conduct their car survey.

LexBob2

I agree that the CR method has flaws, all surveys do. But IMO it's still one of the better, if not the best, sources of this kind of information readily available to consumers.

Much more credible than word of mouth reports from my sister's boyfriend's parents, a friend of a friend etc.

I look at them for things that are of interest or important to me, and make my own decisons. Just one part of the fact finding process.

IS-SV

I agree, that's exactly how I see it.

And I've put the individual model charts to the test by comparing it to actual cars that I've owned and found my experiences closely matched what they were reporting.

I8ABMR

how ever tested I think more ha a few of us have felt the Lexus slip over the past 5 years

Threxx

If CR wants anyone with any clue about surveying methods to take them seriously they need to begin polling the entire general public and not just their own readers. Until then their results are by default largely a self-fulfilling prophecy.

The G Man

That is so true, their polling public consist of their reader only and those readers are buying the cars the CR recommands. You guys see a trend here.

IS-SV

I largely ignore what any magazine recommends (car magazine or consumer magazine or any magazine of any kind). In the case of CR I look at the specific model reliability charts by category (engine major, engine minor, suspension, auidio, etc.), I find this data to be helpful and useful.

And for that reason I'm not bothered by CR's biases.

mmarshall

But reliabilty is one of the big factors in determining what CR does or doesn't recommend. To make the CR recommended list, a vehicle has to pass CR's review/driving tests, have an acceptably good Government crash-safety rating, have an established record of average or better reliability, and, in the case of SUV's and high-center-of-gravity vehicles, pass CR's own rollover tests, which are done with outriggers. Usually, with an all-new or extensively redesigned vehicle, it takes a year or so (and an adequate number of owners filling out the surveys) to start to establish a credible reliability record.....which is why newly-introduced or redesigned models are sometimes not on the Recommended List evrn though previous versions have done well and have been reliable. CR, for example, got burned a few of years ago with the all-new Camry, when they gave it a Recommended rating based in its past record of good reliability. V6 models, though, had defective automatic transmissions, and some older 3.0L V6s had oil sludge/gel problems from not changing the oil enough. CR learned from those experiences....and, now, no longer recommends vehicles based solely on past reliability. They now have to start over and prove themselves with each new model or redesign.