[ducdao]

I would assume the 111 individuals that weren't happy in the survey, they all came from PF

I know those are points, not human but I'm just saying...

[luftfluss]

I would assume the 111 individuals that weren't happy in the survey, they all came from PF

I know those are points, not human but I'm just saying...

Actually, I think it's RH and his 110 other aliases...

[normhead]

Pentax ranks highest in online buyer satisfaction with a score of 899 (on a 1,000-point scale), followed closely by Nikon (891) and Canon (888). Pentax performs particularly well in durability and reliability; variety of features; ease of operation; and shutter speed/lag time, while Canon performs well in the picture quality factor.

Wooooo hooooooo.... first of all, the survey was obviously performed after the cameras were bought, although it doesn't say how long... but everyone had time to use the camera. But canon was first in picture quality.... hmmmm not according to DxO.. and Pentax perfromed best in shutter speed/lag time, you mean like focusing and taking the picture? Not according to almost everyone in the world. I guess those consumers just don't pay attention to the experts.

Funny how if one guys says Pentax sucks, it's gospel, but if a survey says Pentax is doing well, we start saying it doesn't mean anything because of the statistics. You gotta love this site. We have people pushing cameras Pentax doesn't make, we have Canon and Nikon dealers trashing Pentax promoting their wares. And we have people who question every thing that shows Pentax might be OK.

Well, it's really sad buddy, but Pentax might just be OK, and this survey might just show that it's right up there with Nikon and Canon in customer satisfaction. It doesn't say that it for sure because of standard deviation... but it certainly suggests that if you're going to bet, Pentax might be the one to bet on. The deviation thing, affects Nikon and Canon too, so it's equally possible that Pentax is rated lower than they should be and Nikon and Canon are rated higher so the gap is even bigger than the survey show... oops , forgot to completely explain your statistical analysis to the detriment of Pentax... just another dude with a bias.

[Navmaxlp]

But these 5% are equivalent to about 45 points (with a mean of 899 points). And the difference between Pentax and the industry standard is only 11 points or 1.24%. So not very meaningful.

I also did not calculate anything "correctly", I just applied the estimate 'uncertainty = mean/sqrt(N)', and that gives about 32 for mean=899 and N=810 (number of samples). It was more or less a plausibility check only.

coon

I have nothing of value to add to the conversation. I just want to admit that you could have been speaking Greek, Ancient Sumerian or even Alien and I probably would have gotten the same amount of info from it. Not saying it was not useful info, just that I am woefully under qualified to understand more than a few words of it. This went so far over my head, I could have jumped and still missed it.

[mikemike]

Wooooo hooooooo.... first of all, the survey was obviously performed after the cameras were bought, although it doesn't say how long... but everyone had time to use the camera. But canon was first in picture quality.... hmmmm not according to DxO.. and Pentax perfromed best in shutter speed/lag time, you mean like focusing and taking the picture? Not according to almost everyone in the world. I guess those consumers just don't pay attention to the experts.

Funny how if one guys says Pentax sucks, it's gospel, but if a survey says Pentax is doing well, we start saying it doesn't mean anything because of the statistics. You gotta love this site. We have people pushing cameras Pentax doesn't make, we have Canon and Nikon dealers trashing Pentax promoting their wares. And we have people who question every thing that shows Pentax might be OK.

Well, it's really sad buddy, but Pentax might just be OK, and this survey might just show that it's right up there with Nikon and Canon in customer satisfaction. It doesn't say that it for sure because of standard deviation... but it certainly suggests that if you're going to bet, Pentax might be the one to bet on. The deviation thing, affects Nikon and Canon too, so it's equally possible that Pentax is rated lower than they should be and Nikon and Canon are rated higher so the gap is even bigger than the survey show... oops , forgot to completely explain your statistical analysis to the detriment of Pentax... just another dude with a bias.

There is a big difference between how a camera review site such as DxO does their thing and how a market research firm like JDP does theirs and likewise theres a difference in what they are measuring. JDP is measuring customer satisfaction and those customers are basing their satisfaction on their previous camera experiences, which means that they are probably comparing their new camera to their old prior generation DSLR or P&S camera. A review site is comparing peer cameras across brands within the same generation at similar price points with varying degrees of scientific rigor. So the panelists in JDP are not comparing Canon's IQ to Pentax IQ, they are comparing Canon's IQ to their old camera's IQ and canon might be reaching customers who are upgrading from older P&S so those people think it is amazing. Likewise, Pentax might be reaching people who are upgrading from an old camera to a K-5 so their pain points of lag, ergonomics, focus speed, etc... have been addressed and they love it.

If anything should shout out to you that this was done by a marketing researcher rather than a photographer is that "zoom" tops the list of features on a DSLR survey.

[normhead]

Ya, I know, mikemike, the fact that the rules are the same for all manufacturers and Pentax came out on top for some reasons you just made up, that while plausible along with 20 or 30 other possible probablities, are still fiction. So let's look at some of the "facts" I know about. Teh difference between Canon and Nikon and Pentax AF are so close together in good light and low ISO that the average consumer won't be able to tell the difference. That from reading the stats on DxO. This survey is an example of how that plays out in the real world. Consumers can't tell the difference. I was being facetious about the DxO numbers... sometimes you have to know what you're talking about to get my humour.

I have nothing of value to add to the conversation. I just want to admit that you could have been speaking Greek, Ancient Sumerian or even Alien and I probably would have gotten the same amount of info from it. Not saying it was not useful info, just that I am woefully under qualified to understand more than a few words of it. This went so far over my head, I could have jumped and still missed it.

What he said was " Those numbers go against my opinions, in fact they irritate me. Here's why they don't count. Only numbers I agree with count." That should make it easier to understand.

[Gareth.Ig]

But these 5% are equivalent to about 45 points (with a mean of 899 points). And the difference between Pentax and the industry standard is only 11 points or 1.24%. So not very meaningful.

I also did not calculate anything "correctly", I just applied the estimate 'uncertainty = mean/sqrt(N)', and that gives about 32 for mean=899 and N=810 (number of samples). It was more or less a plausibility check only.

coon

Ah. Wrong equation. There are two different questions that need to be answered:
1. Is the Pentax sample size sufficient to make predictions about the entire population? and
2. Is there sufficient information to determine if the rankings between Pentax, Nikon and Canon are "statistically significant" i.e. that you can say one population of customers are in fact happier with their cameras than another based on the data collected.

To answer question 1, you determine the confidence interval based largely on the sample size. If based on your earlier assumption of 810 pentaxians AND an assumption that the sampling is random, the sample mean would be within 3.4% of the population mean 19 times out of 20. Once you get above a sample size of 200, the calculation is fairly insensitive, and you can get to within 5% of population mean regardless of the size of the population.

To answer question 2...talk to a stats prof, all I know is it requires some kind of test called ANOVA and requires more information than we have from the study. Although I think that in this case, yes, it would be unlikely that the differences between Pentax, Nikon and Canon are statistically significant, or at least that I wouldn't assume that they were until someone showed me the numbers.

[normhead]

Although I think that in this case, yes, it would be unlikely that the differences between Pentax, Nikon and Canon are statistically significant, or at least that I wouldn't assume that they were until someone showed me the numbers.

Exactly, however I'd argue , that based on a lot of the bashing that goes on here on this board, I'd argue that the fact that the differences aren't significant is significant. Hows that for some statistical double speak?

[monochrome]

Gareth: To answer your question 2 you would run the same study repeatedly on random respondents and study the variation of the results. Then you might begin to infer that a future study would have a statistically significant probability to result some common parameter set (within your stated error rate).

IOW you might then be able to say with confidence that Pentax cameras have the highest overall user satisfaction rate rather than saying they had the highest rate in one study..

This study is good enough as far as it goes but it isn't predictive, statistically.

However, given what we all know we know about Pentax cameras, it confirms our own common sense experiences.

[lammie200]

To answer question 1, you determine the confidence interval based largely on the sample size. If based on your earlier assumption of 810 pentaxians AND an assumption that the sampling is random, the sample mean would be within 3.4% of the population mean 19 times out of 20. Once you get above a sample size of 200, the calculation is fairly insensitive, and you can get to within 5% of population mean regardless of the size of the population.

I took a few graduate level courses in statisitics. I'll buy that.

This study is good enough as far as it goes but it isn't predictive, statistically.

However, given what we all know we know about Pentax cameras, it confirms our own common sense experiences.

Ditto. Now, what I really want to know is what 4 out of 5 dentists would say.

[Gareth.Ig]

Exactly, however I'd argue , that based on a lot of the bashing that goes on here on this board, I'd argue that the fact that the differences aren't significant is significant. Hows that for some statistical double speak?

Amen brother.

[monochrome]

Now, what I really want to know is what 4 out of 5 dentists would say.

They would say 4 out 0f 5 is too wide a distribution to be useful. Now 95 out of 100 . . .

As securities markets have recently shown us, the real problem isn't the frequency of of unpredicted events it is the magnitude of unpredicted events when they infrequently occur. Tail frequency is relatively predictable. Tail magnitude isn't.

[luftfluss]

Ditto. Now, what I really want to know is what 4 out of 5 dentists would say.

They would say, "Who's your insurance carrier?" All subsequent opinions would flow from that.

[coon]

Ah. Wrong equation. There are two different questions that need to be answered:
1. Is the Pentax sample size sufficient to make predictions about the entire population? and
2. Is there sufficient information to determine if the rankings between Pentax, Nikon and Canon are "statistically significant" i.e. that you can say one population of customers are in fact happier with their cameras than another based on the data collected.

To answer question 1, you determine the confidence interval based largely on the sample size. If based on your earlier assumption of 810 pentaxians AND an assumption that the sampling is random, the sample mean would be within 3.4% of the population mean 19 times out of 20. Once you get above a sample size of 200, the calculation is fairly insensitive, and you can get to within 5% of population mean regardless of the size of the population.

To answer question 2...talk to a stats prof, all I know is it requires some kind of test called ANOVA and requires more information than we have from the study. Although I think that in this case, yes, it would be unlikely that the differences between Pentax, Nikon and Canon are statistically significant, or at least that I wouldn't assume that they were until someone showed me the numbers.

Yep, you are right, I completely skipped question 1 and tried to answer 2 only. If you then assume that everybody gives a rating = Pentax population mean + noise, then your standard error of the (sample) mean is given by the formula stdv(noise)/sqrt(N), if you have normal distributed noise (bell curve). Now we don't have normal distributed noise(=varying buyers experience) here, that's why I wrote "plausibility check". With respect to distribution independent tests (I assume the one you mentioned is one of those, I don't know it), I have no experience. But they all should give uncertainties in the same order of magnitude, if you don't have large outliers (and they couldn't vote <0 or >1000).

That much about theory. To my shame I must confess that I used 899 (which is the mean) for the standard deviation in my calculation, and this was indeed wrong. So to turn the argument around, having a statistically meaningful difference of 11 points in the mean for 810 (10% of the buyers) samples, the standard deviation in the noise must be significantly smaller than about 11*sqrt(810)=313. This is already the case for a uniform distribution between 0 and 1000 (stdv=289), probably not many people even voted with less than 500.

So yeah, right formula but wrong numbers...

coon

[LFLee]

I thought only we Pentaxians are like brand die hard fans....
but look at all the online articles that report this survey result.... Nikon and Canon fans are getting really upset about this survey. LOL

Good publicity for Pentax, but do nothing good on my photography skills... I just hope Pentax lens prices are going back to what it used to be...

Come on, Pentax! now beat the Canikon on the FF segment!!

Lee