Results of M8 Lockup/Failure Survey

[stunsworth]

Advertisement (gone after registration)

But we aren't sampling the whole population John, we're just sampling the ones who responded to the question. I agree totally that if the entire population were to be sampled we would be able to draw meaningful conclusions about the failure rate.

 

I'm really not wanting this to sound like a 'bashing' of Sean who I have a great deal of respect for, all I'm saying is that the survey doesn't give us an indication of the true failure rate.

[larry]

But we aren't sampling the whole population John, we're just sampling the ones who responded to the question. I agree totally that if the entire population were to be sampled we would be able to draw meaningful conclusions about the failure rate.

 

I'm really not wanting this to sound like a 'bashing' of Sean who I have a great deal of respect for, all I'm saying is that the survey doesn't give us an indication of the true failure rate.

 

 

Steve,

 

Sean doesn't claim that his numbers are exact or meaningful in a global sense. You should read his original post carefully. I think you may be trying to define the results in a way that he's not.

 

The company I work for conducts surveys like this to glean information on trends, but doesn't make marketing decisions based on them. That's a function of much more extensive testing, however these small surveys still have value as a "weathervane" of sorts.

 

Larry

[larry]

You could consider the active M8-owning membership of this forum to be a population. Most didn't come here to report problems; most came here to discuss the M8, and I'd argue that most seem to have a tendency to play down problems.

 

 

John,

 

That's a bit of an assumption, why should I (or anyone else) report problems I haven't had?

 

Larry

[Riley]

Out of a random sample of 400, recording the number of failures and non failures, you could be 95% confident that the number of failures is a true reflection. Less than 400 is just a less confident appraisal. 140 is a useful quotient

 

In this study, which although short, isnt a random sample, indeed most of the respondents would more often than not, have had a failure. So 10% is nothing like the reality.

 

That said, it is less than what it seemed to me, and it is a respectable effort by Sean to get the numbers.

[sean_reid]

Steve,

 

Sean doesn't claim that his numbers are exact or meaningful in a global sense. You should read his original post carefully. I think you may be trying to define the results in a way that he's not.

 

The company I work for conducts surveys like this to glean information on trends, but doesn't make marketing decisions based on them. That's a function of much more extensive testing, however these small surveys still have value as a "weathervane" of sorts.

 

Larry

 

Exactly! Thank you. The survey is what it is and I was careful to point out that it could not be generalized to a larger population. People who choose not to read my text are bound to draw some incorrect conclusions.

 

What the survey does tell me is that this is an issue that deserves Leica's attention. 13 out of 140 is not ideal.

 

Cheers,

 

Sean

[Jamie Roberts]

{snipped}

If, for example, you want to take an even smaller population, professional photographers who have frequented this forum since before the camera was released (i.e. before you could complain about problems), then, as close as I can tell, 100% of them have experienced the problem. That doesn't mean that 100 percent of all professional photographers have had the problem, but it does mean that of those you know, a more than expected number have had the problem. (With that small a sample, with Leica-level QC, you might expect the possibility of a defect, but you wouldn't expect that all of them would see one.)

 

JC

 

But John, even that's wrong, since I'm a pro and haven't experienced a lock up or DOA or even the need to reboot.

 

The problem actually--though I'm not going to beat this to death--is that this is qualitative information and not, strictly speaking, quantitative, unless you restrict your assumptions to something like "a percentage of the people who responded to the survey" which, unfortunately, has no causal or general applicability whatsoever. None at all.

 

Now, there is *good* qualitative information, which says there may be (emphasize may) be a common problem with the M8 and its manufacture or its programming. The amount of people complaining only verify that statement, and only that statement.

 

That's absolutely valueable information, but it's more of a "question" than an "answer."

 

But a good question is a valuable thing; then you can create the proper quantitative analysis to understand how widespread this might be, or the engineering analysis to see if there really is a problem (or where the problem lies).

 

Without Sean's work, Leica would have less information about the quality of the M8 experience for the people that reported, and wouldn't be able to go do the hard quantitative work. This is a good and great thing, even if it may happen only to a very, very few customers overall, or is caused by different underlying problems.

 

(But this is why numbers are dangerous in qualitative information; they look like "hard causal data" but, of course, they aren't).

[wparsonsgisnet]

Advertisement (gone after registration)

Way too much coffee, folks.

 

Sean made is pretty clear -- and his writing is customarily clear -- that this was a survey [my word].

 

What we have going for us here is that Sean and several others on this forum have the ear of Leica. Not so bad.

 

Sean's results should be just what Leica wants: an itemization of failure problems and, perhaps, circumstances. Only Leica has the data to decide what part of their production these failures represent.

 

I suspect that Sean went to a lot of work on Leica's behalf, just to be a nice guy. Not so bad.

 

Regards,

[John Maio]

<snip>

I'm told that Leica is investigating this concern and I await further information from them about what solutions are possible.

<snip>

 

.

 

Sean,

This implies that they don't have a solution yet. Does this mean your camera was not repaired as yet?

 

If it was, what did they do to fix it?

[Guest guy_mancuso]

Way too much coffee, folks.

 

Sean made is pretty clear -- and his writing is customarily clear -- that this was a survey [my word].

 

What we have going for us here is that Sean and several others on this forum have the ear of Leica. Not so bad.

 

Sean's results should be just what Leica wants: an itemization of failure problems and, perhaps, circumstances. Only Leica has the data to decide what part of their production these failures represent.

 

I suspect that Sean went to a lot of work on Leica's behalf, just to be a nice guy. Not so bad.

 

Regards,

 

Thank you Bill ,said this twice already the numbers are meaningless the issue is what counts and that data gets sent to leica to find solutions . Bottom line , end of the day that is all that matters . Now i am going for more coffee and go fix the pool. Than process several thousand more images to deliver to clients. BTW i had a dead camera too, so not avoiding the issue here and leica certainly knows what happened to mine and it is back and is working like it is supposed too.

[sean_reid]

Sean,

This implies that they don't have a solution yet. Does this mean your camera was not repaired as yet?

 

If it was, what did they do to fix it?

 

It doesn't imply that, actually. They sent me a different camera and I don't even know if they've looked yet at the camera I sent in. If I get info. I'll pass it on.

 

Cheers,

 

Sean

[sdai]

I'm assuming that everyone with a flat dead M8 "has to" send the camera back to Solms, right? so Leica must already have the exact numbers.

[sean_reid]

New paragraph in article draft, for those who are interested:

 

"I contacted Leica to discuss the results of the survey and to get some information about the total number of M8s sold to date and the percentage returned to Leica because of electronic failure. Unfortunately Leica, like many other companies, has a policy of not releasing these statistics. They were able to say, however, that the percentage of cameras returned (because of failure) was much lower than 9 percent and that Leica would certainly consider a 9 percent failure rate to be unacceptable. Again, we must remember that a sample size of 140 is small and that this survey was not random."

[Riley]

I'm assuming that everyone with a flat dead M8 "has to" send the camera back to Solms, right? so Leica must already have the exact numbers.

 

that gives the numbers for that condition Simon

there are other situations considered in the poll

but in a sense you are right

you could with the knowledge of both data sets

extrapolate the data into a reasonable estimate

[wparsonsgisnet]

New paragraph in article draft, for those who are interested:

 

"I contacted Leica to discuss the results of the survey and to get some information about the total number of M8s sold to date and the percentage returned to Leica because of electronic failure. Unfortunately Leica, like many other companies, has a policy of not releasing these statistics. They were able to say, however, that the percentage of cameras returned (because of failure) was much lower than 9 percent and that Leica would certainly consider a 9 percent failure rate to be unacceptable. Again, we must remember that a sample size of 140 is small and that this survey was not random."

 

Exactly as one would expect. The important thing here is to continue to reach Leica's ear.

 

Nice going, Sean.

[Jamie Roberts]

that gives the numbers for that condition Simon

there are other situations considered in the poll

but in a sense you are right

you could with the knowledge of both data sets

extrapolate the data into a reasonable estimate

 

No, you couldn't actually--the survey here would still be self-selecting and not representative of M8 users as a whole.

 

Though I supposed it depends on your definition of "reasonable."

 

The *numbers* are meaningless. I wish Sean wouldn't even mention them, in a way, because, as I've said (and others have too) the facts of the failure are more important than their incidence.

[Riley]

Jamie

I still feel the 'reboot/camera' died relationship could hold

the 'no fail' end of the data set I interpret should be larger

so with Leica having there own 'camera died' and clearly more reliable stat, that of returns v/s total population

you could extrapolate the reboot stat, that they wouldn't ordinarily see

[sean_reid]

No, you couldn't actually--the survey here would still be self-selecting and not representative of M8 users as a whole.

 

Though I supposed it depends on your definition of "reasonable."

 

The *numbers* are meaningless. I wish Sean wouldn't even mention them, in a way, because, as I've said (and others have too) the facts of the failure are more important than their incidence.

 

Jamie?

 

The numbers are not meaningless but they must be related to the given sample. Its impossible to present this information well without the numbers but I count on the readers to read my text carefully and look at the numbers in the correct context.

 

Sean

[carstenw]

Well, the numbers are what they are, as you say, but since the set was self-selecting, they don't prove anything at all. All we know is that of the 140 who answered, 13 had their M8 die, nothing more. Nothing about the general set at all.

 

If you think about it, this makes sense. For example, it is likely that people who had their M8 die on them are *much* more likely to answer, to make sure that their voice was heard.

 

For this survey to be statistically valid for the purpose of saying anything about the entire set of M8 owners, the people answering should have been chosen at random from the entire set of M8 owners, not self-selected from M8 owners who read this forum, and happened to be here on the dates in question.

[Jamie Roberts]

Jamie?

 

The numbers are not meaningless but they must be related to the given sample. Its impossible to present this information well without the numbers but I count on the readers to read my text carefully and look at the numbers in the correct context.

 

Sean

 

Sean--I completely agree with you, and understand the quandry of presenting the information without the numbers. What I mean is they're meaningless to estimating outside the sample (which was what Rob was saying).

 

And because people will jump to the "we can estimate a trend here" conclusion, well, I wish there was a better way to do it, is all. Please don't use graphs, either--they might be even more confusing

 

Just me Again, I think this is really important information for Leica to have.

 

I guess if you were being strict here, you could say something like the survey pointed out that "within the core group of responders, problems were "triangulated" (in other words, confirmed well beyond a single incidence) and may point to issues of manufacture control or software control.

 

Yikes. It's a long time since I did this kind of stuff (but I did do it )

 

Sorry again--I said I wouldn't beat it to death and now I have I'll be quiet now!

 

I should have just said "Amen" after Bill's post!

[Guest guy_mancuso]

Couple Hail Mary's and all sins are forgiven. at least that is what they tried to tell you. LOL

[Advertisement]

Hallo,

Look here: Interesting Leica gear