Mazda3 reliability - FMVperformance.com : The site for all your Ford Mazda and Volvo needs
Reply
 
LinkBack Thread Tools
post #1 of 30 (permalink) Old 10-23-2007, 11:30 AM Thread Starter
Short Track Racer
 
Join Date: Nov 2005
Location: Michigan
Posts: 229
Mazda3 reliability

Because existing sources of vehicle reliability information left much to be desired, in late 2005 I started conducting my own reliability research at TrueDelta.com. TrueDelta reports absolute stats like "trips to the shop" that make the differences between cars much clearer. Results are updated four times a year, so any significant changes in a model's reliability become apparent quickly.

The Mazda3 has been one of the best represented models, with over 230 owners already signed up (plus another 140 Proteges). But more participants would still be helpful. The larger the sample size, the more accurate the results and the more detail I can provide.

Participants simply report repairs the month after they occur on a one-page survey. When there are no repairs, they simply report an approximate odometer reading four times a year, at the end of each quarter.

To encourage participation, panel members will receive full access to the results for free.

For the details, and to sign up to help out:

Vehicle reliability research


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.

Price comparisons, quick and thorough
More useful reliability information
mkaresh is offline  
Sponsored Links
Advertisement
 
post #2 of 30 (permalink) Old 10-26-2007, 10:24 AM
Moped Rider
 
Join Date: Aug 2007
Location: Indy
Posts: 42
Re: Mazda3 reliability

Interesting setup. I joined.

But, they need to quadruple the membership to have statistically relevent data.
Hopefully they will.....
quicksilver is offline  
post #3 of 30 (permalink) Old 10-26-2007, 12:46 PM Thread Starter
Short Track Racer
 
Join Date: Nov 2005
Location: Michigan
Posts: 229
Re: Mazda3 reliability

Thanks for signing up. One thing that really gets to me are people who say, "I'm not signing up because the sample size is too small." Very happy to see you're not one of them.

That said, I am not sure what you mean about the sample sizes being too small to be "statistically relevant." The current sample sizes are large enough for ballpark results. They're not large enough for precise results, but precise results are not necessary to sort out which cars are reasonably reliable and which are not.

I've made correct predictions based on as few as a dozen responses: last August I said the repair rate for the V6 Camry was running double that of the four-cylinder. Two months later it's big news when CR says the same thing.

With the Mazda3, the results for all four model years are nearly the same. I don't think this is a coincidence.

Would I like the sample sizes to be larger? Absolutely. With more data I could go into more detail, and more precision is always better. And the sample sizes are getting there. In the November results the sample size for the Nissan Versa is over 120 and a few others are close to 100. The panel is currently growing by about 1,500 vehicles per month, including 35 so far today.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.

Price comparisons, quick and thorough
More useful reliability information
mkaresh is offline  
post #4 of 30 (permalink) Old 10-26-2007, 02:27 PM
Grand Am Series
 
Join Date: Jul 2007
Location: Daphne, AL
Posts: 253
Re: Mazda3 reliability

Signed up

The earth travels through space at 66,000 miles per hour.
stranger706 is offline  
post #5 of 30 (permalink) Old 11-05-2007, 03:41 PM
Moped Rider
 
Join Date: Aug 2007
Location: Indy
Posts: 42
Re: Mazda3 reliability

"I am not sure what you mean about the sample sizes being too small to be "statistically relevant." "

Don't take it as a knock....
It is a basic rule of statistics that a standard distribution can not be predicted accurately with less than 30 data points for a specific subset. Perhaps more, if current data is influencing new responses.
But, 30 is usually a minimum allowed when doing statistics.

With 300 data points, you can usually, accurately predict out 3 standard deviations or more.
(that covers more than 99% of a population)
That raises eyebrows from engineers because now data has become "statistically" relevent.
You have data that accurately represents an entire population.

If you have 300 owners of a specific car responding, and 30 of them have the same complaint, it is fair to say that 10% of everyone with that car has the same issue.

That's what I meant. Yes, many subsets appear to be filling in nicely.
Yes, with small subsets, you can catch a lot of catastrophic stuff or get lucky.
You can also be totally misleading.

As more and more models get more than 30 data points for each year or model generation, more and more people will be willing to benchmark off this data. When lots of populations of subsets start to fill out, with more than 300 data points, even automotive engineers will start paying attention.

After the data has grown to this point of maturity and develops a track record of integrity over time, it becomes something that is quoted and used for decision making.

Here endth the lesson... 8)
quicksilver is offline  
post #6 of 30 (permalink) Old 11-05-2007, 09:21 PM Thread Starter
Short Track Racer
 
Join Date: Nov 2005
Location: Michigan
Posts: 229
Re: Mazda3 reliability

Thanks for the explanation. Unfortunately, I'm too tired to properly respond to this at the moment.

It will have to suffice to say that the current results are neither accurate to three standard deviations, nor do they rely on luck.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.

Price comparisons, quick and thorough
More useful reliability information
mkaresh is offline  
post #7 of 30 (permalink) Old 11-14-2007, 01:08 PM Thread Starter
Short Track Racer
 
Join Date: Nov 2005
Location: Michigan
Posts: 229
Re: Mazda3 reliability

Just a bit more of a response:

--the difference between 25 and 30 is negligible, though of course you must draw the line somewhere

--precision increases with the square root of the sample size, so 100 is twice as good as 25, and 225 is three times as good

--100 is CR's minimum, and 225 is JD Power's target, but they also try to provide results by subsystem, which I don't at this point; so for what I do my sample is at least as sufficient as theirs are for what they do

Would I like larger sample sizes? Absolutely, and working very hard to get them. But are they necessary for useful information? No.

It's not just dumb luck that the results for all four years of the Mazda3 are close to one another.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.

Price comparisons, quick and thorough
More useful reliability information
mkaresh is offline  
post #8 of 30 (permalink) Old 12-05-2007, 02:44 PM
Moped Rider
 
Join Date: Aug 2007
Location: Indy
Posts: 42
Re: Mazda3 reliability

Interesting post.

"--the difference between 25 and 30 is negligible, though of course you must draw the line somewhere"
fyi,
I believe the 30 piece number is the minimum required for a 95% confidence of an assumed standard distribution. Statistical confidence drops rather sharply as the sample size goes down from 30. Conversely, it grows rather slowly as it gets bigger (the 300 number is 99%).
So, it's just a "most bang for the buck" number.

For this reason, the 30 number was adopted by Shanin quality techniques for iso-plots and such. (as well as Six-sigma folks and everyone else)
As with every rule there are plenty of exceptions. Data is only as smart as the people using it.

If you poke around with Google for "central limit theorem" and "Shanin",
You can dive into it as deep as you like. (assume you have trouble sleeping at night )
Suffice it to say that quit a few really, really, smart folks say that's where the line should to be drawn.

The other side of the coin is: you are correct about subsystems or families of data.
If you are not looking for a specific defect and are willing to look at broader families of data (multiple years or multiple subgroups), high failure rates do not need as much data to accurately reflect general quality.

You can pick any number you want as a target sample size, but as you said, the more you capture, the more you will see families of data. Example: cars made in June/July often stuggle more because this is the normal timing for model launches and model year change issues. I have not seen anyone try to publicly capture that data before. That would be a cool one to add into the mix of rankings.

quicksilver is offline  
post #9 of 30 (permalink) Old 12-05-2007, 05:39 PM Thread Starter
Short Track Racer
 
Join Date: Nov 2005
Location: Michigan
Posts: 229
Re: Mazda3 reliability

Based on the way confidence intervals are calculated, the width of these intervals varies inversely with the square root of the sample size. There's no reason for a big difference between 25 and 30. It's not like you have crap at 25 and then a solid result at 30, five more people don't make that much of a difference.

I don't formally pool sample sizes, but do look for patterns or a lack thereof within the entire population of results, and leave comments accordingly. For example, the results for four years of the Mazda3 are nearly identical. So I'd say that they reinforce one another.

Believe me, if all of my sample sizes were 30 or even 50, I wouldn't mind, not one bit

But I have learned that the way to get to those sample sizes is to start releasing results at the lowest level that provides useful information. At 25 and even at 15 it's possible to gather useful information from these results. I base this on the general stability of these results from quarter to quarter, from model year to model year, and when comparing my results to others'.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.

Price comparisons, quick and thorough
More useful reliability information
mkaresh is offline  
post #10 of 30 (permalink) Old 12-14-2007, 01:01 PM Thread Starter
Short Track Racer
 
Join Date: Nov 2005
Location: Michigan
Posts: 229
Re: Mazda3 reliability

Updated results in February. Additional participants always helpful.

Vehicle reliability research


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.

Price comparisons, quick and thorough
More useful reliability information
mkaresh is offline  
Sponsored Links
Advertisement
 
Reply

  FMVperformance.com : The site for all your Ford Mazda and Volvo needs > Misc > Lounge

Quick Reply
Message:
Options

Register Now



In order to be able to post messages on the FMVperformance.com : The site for all your Ford Mazda and Volvo needs forums, you must first register.
Please enter your desired user name, your email address and other required details in the form below.

User Name:
Password
Please enter a password for your user account. Note that passwords are case-sensitive.

Password:


Confirm Password:
Email Address
Please enter a valid email address for yourself.

Email Address:
OR

Log-in










Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page



Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On

 
For the best viewing experience please update your browser to Google Chrome