BMW X5 (E53) reliability - how does it compare?

[mkaresh]

Ten more have signed up, and some model years are now 3/4 of the way to the minimum.

Vehicle reliability research

[weirluo]

don't think it's going to be a fair result,

most first hand X5 owners don't spend much time on car forums

from my market research experience

a phone survey to randomly sampled owners would be much more better

[mkaresh]

The research design largely compensates for the lack of a random sample by having people participate continuously, beginning with when they sign up.

Would I prefer to survey a random sample of vehicle owners on a monthly basis? Sure, if the response rate were high--which it wouldn't be. And even if the response rate were 100 percent, no one would pay the resulting cost of the research--such a survey would cost millions.

What I have is a method that yields surprisingly robust results, with low enough costs that I can provide the results for free to anyone who participates.

[weirluo]

Originally Posted by mkaresh

The research design largely compensates for the lack of a random sample by having people participate continuously, beginning with when they sign up.

Would I prefer to survey a random sample of vehicle owners on a monthly basis? Sure, if the response rate were high--which it wouldn't be. And even if the response rate were 100 percent, no one would pay the resulting cost of the research--such a survey would cost millions.

What I have is a method that yields surprisingly robust results, with low enough costs that I can provide the results for free to anyone who participates.

not interested in challenging your bussiness, but low cost results might be misleading results too.

[mkaresh]

I feel the proof is in the results themselves. A few seem off, and I've commented them as such. But this is more a matter of the small sample sizes than the source of the sample.

Here are the results for the BMW 3-Series, in successful repair trips per 100 cars per year, with an * if the sample size is small, and the average odometer in ( ):

2000 200* (119600)
2001 135 (73000)
2002 151* (71500)
2003 104* (56500)
2004 75 (41200)
2005 69* (33900)
2006 68 (24900)
2007 52 (10200)
2008 31 (7300)

Though half of these years have sample sizes below my minimum, we still get a nice clean curve, especially from 2002 on. The repair rate for the 2000 seems a bit high, and that for the 2001 a bit low, but the odometer readings likely account for this--those 2000s have a lot of miles on them. Did the 2000 MY begin early for the E46?

[weirluo]

again, you don't get my point, no offending, but i am a quantitative analyst with a math PhD degree, used to work in a market research firm, so if my firm asked me to do the research, I would do it in a proper way, such as being equipped/backed by statistics and sampling theories and covering all customers rather some a small portion, in this case the small portion only includes those who get on the internet and come across ur website.

Some of your results could appear correct, but some of them may be misleading. you can't use part of your results to evaluate/proof your methodology.

Nevertheless I get where you come from, at least you can demonstrate some results, not always right, but perhaps better than no result.

um... do I sound like a nerd, again?????? Damn, stop it!!

[mkaresh]

You sound fine to me, and I get your point.

The key thing to realize, and that often escapes people at first glance, is that this isn't an opinion poll. With an opinion poll it is critical to have as representative a sample as possible.

In this case, it's important to ask how the people who are being sampled are likely to differ from the general population, AND how this might affect the results.

Ways forum visitors might differ from the general population:
1. Drive more aggressively
2. More likely to mod
3. Might maintain their cars better
4. Might notice a problem with their car only after hearing of it on a forum
5. Came to the forum because they have been having problems

Of these, I'm most concerned about #3 and #5. But looking at the details of the repairs that are reported, the great majority, including the many electrical problems people report:

1. Could not have been caused by how the car was driven
2. Could not have been caused by a mod (and respondents are asked if their cars are heavily modded--few are)
3. Could not have been prevented by maintenance--there's no way to maintain a window regulator, for example; that said, maintainable parts do fail more often with older cars, so this becomes more of a factor with those
4. Are problems that anyone would notice
5. Occurred after the person learned of the research and signed up

Yes, there might be some distortion from the above factors, but not enough to render the results misleading unless someone is trying to split finer hairs than I'd suggest. This isn't a presidential election poll, where a result that is off by more than a couple percentage points would be misleading.

[mkaresh]

124 X5 owners signed up to participate so far. A good start, but more are needed.

Vehicle reliability research

[motordavid]

I participate in your research forum, but I am surprised that 2 of
my cars: '01 X5 and '02 Vette are not in the hunt, due to lack
of respondents for those years/car(s).

NEway, our CRV-Fridge makes the list.

The reliability report is pretty interesting for the cars "enrolled", imo.
BR,mD

[Quicksilver]

Supergreg wrote about this thread more than a year ago
as noted below. So I'm asking can somebody help me out
with this thread today?

How can this person continue to sell a product under the
guise that there will be some benefit to X5world but never
produces anything we can use?

I'm viewing this thread as a free site advertising medium.
I believe this person and his company needs to at least follow
the rules for selling on this site.

Am I wrong about this????

Originally Posted by SuperGreg

I have a bit of an issue with this thread. While I applaud your research and willingless to share information with our X5 group (at least that is how it has been served up), the issue I have is that the only way to access your research once you complete it for the X5, is to A) Become a panel member B) Pay $24.95 per year for the access rights. In addition, I believe that with your minimum sample size of 25 (n=25), this would represent only about 1/10th of 1% of the 2007 X5 population of 23,798, and would not be statistically significant, even at the absolute lowest confidence levels i.e. in order to achieve a confidence level of .95 or a margin of error of .05 with a probability of .50, your sample size would need to be > 300. I'm viewing this thread as a free site advertising medium for you. I understand that you have, "a long time horizon and no investors to pay off". That makes sense to me since as long as this thread exists, it's free advertising for your site. I'll leave this thread active until you respond, but if you want the thread to remain active, you will need to give our members a way to access the information free of charge. Otherwise, this thread represents zero benefit with respect to information for our members, and that is the purpose of our forum . Now, we can always discuss your becoming a forum sponsor. Please send me a PM for further details .