I also forwarded the info to a couple of local guys who have owned two separate RX-8's over several years with replacement engines.

Also, you may want to add "new OEM replacement engine" as an option because when the failures first started showing up a couple of people actually got new engines from Mazda instead of remans.

geez 9k, stop n00b'ing up the survey! :fruit:

I hesitate to add that. Mainly because there are alot of people that don't know the difference. I think the error rate it would introduce would be larger than the error rate for not knowing how many of each.


Fixed :lol2:

Okay, as you know I filled out the test one, but I just PM'd you with my E-mail addy and request for an engine ID, so I can fill it out for real.

Seems like you used at least of a few of my inputs or thoughts and others, so I wanna see how it turned out. but will wait for the ID. Okay, maybe I'll take a quick peek and see how far I can go without the ID. I'm sure I'll just get to see the first page.

Dave


ID recieved...........filling it out now!

Valium works good for this! :wavey:

Don't jack it up with all kinds of really "unknown" data before it gets off the ground. (just saying)

P.M.ed for ID's.
I don't know exactly were my last 2 engines came from. I can make a educated guess by what I know from this site and the time of replacement.

Yup, if you don't know exact numbers, guess are fine, as long as they are reasonable :)

:lol:, sorry!!! Gosh!!

You are right! I failed statistics in College :lol:

Should be easy to get since I manage a chain of Psychiatry clinics :lol:

This must have taken a lot of work on your guys part. Excellent job! I'll PM you and try to fill it out between tonight and tomorrow. I have to search around for my service records when I get home from the office.

Just passed out another 10 IDs.

Thanks guys!

I'll get some of the results up tonight. Not that they will mean much yet, but just a taste of what kind of data we will be getting.

Just completed mine................hope I get an "A".

FYI...........I see you asked the question about different types of intakes, but right before that you asked for cooling related mods(fans, lower temp options on fans, etc).
I don't think it's really needed, but should there be something about cold ram air intakes, to differentiate between just sucking up hot engine air?

^ Sweet. I am curious to see what the results are after a significant amount of people post up.

I started to reply, 'do you think that makes a different in core heat build up?' But I fell back to 'lets let the data tell us'. :)

I wasn't really thinking about that aspect, as that is in the 'power mod' section, as more power generally means more heat, even if faintly. The AEM/MS intake is split out because it is one with actual measurable gains, even if small, where as 'other' generally don't. Most aftermarket intakes generally improve the airflow after the radiator though, which is the main reason for including them at all :)

Ill add it in as soon as I can safely make a change without causing an error for anyone currently in it.

Looking at the account settings, they did increase the limit to 450 responses, but deleting a response doesn't remove that mark against the account :(

So we are at a current limit of 448 engines.

So I am going to block the "going back and fixing", since that counts as another response. (hence dropping 2 from 450)

http://www.racingbeat.com/FRmazda4.htm

Not going to make me popular, but I trust this coupled with the ram air duct more than the AEM/MS. That's just an opinion............again, not sure it leads to engine longevity in any way what-so-ever.


I'll shut-up now!

We might have to eventually ask for cash, so we can get more than 450 repponses! Let's see how fast you fill those up!

I look forward to taking part, but you do realize that you can only use the results of this survey to draw conclusions about RX-8 engine reliability of the people who elect to respond to the survey. In other words, you cannot use the results of this survey to suggest the reliability rate of Renesis engines at large.

To do that would require drawing a random sample of all RX-8 owners, which this survey clearly can't do. So you're using a non-random sample that's not representative of all RX-8 owners.

In other words, for argument's sake lets say that there are 100,000 RX-8s in the world, and that 2,500 of them have had engine reliability problems. That would mean that 2.5% of Renesis engines are not reliable. If you were to randomly select a sample of those 100,000 owners, statistics would tell you the incidence of unreliability is 2.5%, + or - a small percentage, which is sampling error.

Now let's say that there are 2,000 members on this forum. And let's say that of those 2,000 members, 1,500 respond to the survey. And let's say that, of those 1,500 members who respond, 500 have experienced engine problems. The survey would then suggest – at least, to people who don't know statistics – that 33% of RX8s have engine reliability problems. A far cry from the actual 2.5% incident rate.

The results will be misleading because there's a much higher probability that RX-8 owners who experience engine problems will find their way to RX8Club.com. Why? People with problems go online seeking information and solutions; people without problems are less likely to do so. Since this survey is not based on a random sample, you can't use it to generalize about all RX-8s out there.

Maybe this definition from wikipedia will help:

"Nonprobability sampling is any sampling method where some elements of the population have no chance of selection (these are sometimes referred to as 'out of coverage'/'undercovered'), or where the probability of selection can't be accurately determined. It involves the selection of elements based on assumptions regarding the population of interest, which forms the criteria for selection. Hence, because the selection of elements is nonrandom, nonprobability sampling does not allow the estimation of sampling errors. These conditions give rise to exclusion bias, placing limits on how much information a sample can provide about the population. Information about the relationship between sample and population is limited, making it difficult to extrapolate from the sample to the population.

Example: We visit every household in a given street, and interview the first person to answer the door. In any household with more than one occupant, this is a nonprobability sample, because some people are more likely to answer the door (e.g. an unemployed person who spends most of their time at home is more likely to answer than an employed housemate who might be at work when the interviewer calls) and it's not practical to calculate these probabilities."

Anyway, just thought this should be clear.

^.......don't think too much!
Just looking for possible trends is all I think we can really hope to glean out of it.
He's got at least one squeaky clean data point. That I do know.

Duly noted. But saying people who visit online forums only do so in search of resolving problems in not accurate IMO. Online car forums are a part of being a car enthusiast IMO.

LOL. I'm actually kind of a silly, fun kind of person in real life although, reading what I just wrote, I can see how that might be hard to believe. Thing is, I was a psych major forced to study statistics against my will.

I work in the Engineering field and and humping on statistics quite a good portion of the day, so I know the drill. I guess I was just pointing out that trying to capture the data, putting it into some reasonable format, spitting it out, and seeing what it says could just be interesting.......nothing more. :uhh:

I actually come on this site to get away from work and give the cranium some rest!
Sometimes that actually costs me as I tend to let go so much, I end up brain-farting! :sadwavey:

^ Yeah, it will be interesting in a lot of ways. Can't wait to see it. (And thanks, by the way, to everyone who's gone to alotta trouble to set the thing up.)

I 100% understand. I am a data analyst by trade, regularly dealing with >3 TB of raw data from across the country, looking for variances, exceptions, etc... I aslo use this same survey system for more raw data collection, and understand the issues around data collection from an error-prone human being. :)


I recognize that this is only a sampling of:
- People that own an RX-8
- People that are internet 'savy' enough to visit a forum board
- People that see the survey
- People that choose to participate

And because of that, there will be a bias caused by:
- People that have issues are more likely to come here
- People that are more likely to recognize the value in this have some inherent inclination to take better care of their cars
etc...

I know it isn't a true sampling, and the only people that can get a true sampling is Mazda, and if Mazda took over we would be lucky to see the results, even if they left it running. I am high enough up the corporate ladder to be involved in liability concerns, politics, shareholder concern, etc..., so I'm not naive to what would happen if Mazda took it over.

That being said, if I am no where near my cap on responses, I plan on tucking cards under windshield wipers of RX-8s at Sevenstock (which adds another bias into play, of both regional data as well as the 'hard-core' quotient). Then there is seeing if I can get it sent to the RX-8 owner list of TrueDelta, mentioned on AutoBlog or another similar sight, etc...


The reason I decided to do this, is that even if it isn't a statistically true sampling, there is anything else that comes close to holding a candle to this detail about our engine that is available to the general public. If Mazda finds out about this and applauds it to the point of assisting, and is willing to leaving it open to the public, that would be an ideal scenario. If they offer assistance, but with the requirement that they hold the data and the release of the data, then the project would fail, even if you could get people to respond to Mazda with what exactly they are doing to their 8s.

Mazda is welcome to the data from this, as I will not provide any identifying information on who's engine is being referred to for any 1 line. If it helps to make a better rotary, and if some tid bit of information from this survey makes them look up and say "oh, we need to do THAT" or even "we need to test THIS!" then I think that this project would have been successful beyond any expectations.



I am just getting sick of all of the theorycrafting and opinions with basically NO data to back it up. So rather than complaining about it, I'm doing something about it, and gather that data.

Wait a minute. You typed out all that, but didn't bother asking for an engine ID to participate!!!!! :eek: :nono:

:lol2:

While I really can't confirm or deny it, Mazda has been known to visit this site on more than one ocassion.....and quite frankly...........probably does so on a semi-frequent basis.

I have close to 130k Mazda can look at my post all day long I have emailed the OP. This info will be great to have around in the years to come.

This is what I like to call future planning.