Attached Link: Survey Results Download Link

Here are the results, presented in tab delimited format ready for setting up in a database. Zip file. Some observations: Poser 6 is now the preferred program by over 50% of the users. Poser 5 is number 2 DAZ studio is number 3 (but not by much, so it's a toss up). All other versions of poser consist of less than 10% of the preferred userbase, combined. This is inclusive of Poser 3, 4, and Propack. Victoria 3 is the most preferred figure. Victoria 3 is the figure for which the most is still wanted. Victoria 3 is the figure for which there is too much in the marketplace. Aiko is the #2 figure. There is a HUGE demand for Male clothing for all the figures. The least popular figures are Don and Judy. There's a tremendous amount of data available in these results. GOing through it requires patience, time and effort, as well as an interesting understanding of the way things work. There are 609 effective responses in the final break, and the final accuracy level is approximately +/- 7.3412%. The next survey results will be presented in a db style format. THank you to all who repsonded and participated -- the survey is an incredible snapshot of the community as a whole, and should begin the work of expanding botht he community and the market and enabe us all to start seeing things we really want. Thanks again to everyone. I'm off to see what I can do about actually *using* this info, lol....

It's only an informal survey on poser, not a national security survey done by the FBI. Relax :)

Message edited on: 02/13/2006 23:32

Attached Link: http://poserpros.daz3d.com/

> Quote - isn't there a daz freebie you should be promoting?

Good point!

Poser Pros has some nice freebies on their front page this week. The V3 Teddy texture is quite cute :)

Message edited on: 02/13/2006 23:43

Attached Link: http://www.oddditty.biz/Survey/Questions.html

Although I'm normally disinclined to respond to rudely phrased requests, I'm going to provide a "dead" version of the quiz again for reference. The capability of the end user is not something I can control. Everything is provided. not retarded. ynsaen. Significant difference...

Thanks ysaen, it looks like an interesting read. I didnt find any problems getting into the data and getting some results, not sure why you'd need a common key. I'm guessing the IP add is largely useful for grouping the results by submission so overall would be the only key available. Otherwise you can get information from the individual tables. Interesting to see that nearly 70 of people dont use the CP figures but most people love P6 most. One of the brilliant things about P6 is the figures, especially James. John

There are some really incredible results in there, John -- I've been limited while awaiting repair of my main system to using open office..org and peeking in at the data and some of it is just absolutely astonishing, lol. Some things I expected, but most of it is a total surprise to me!

It's cool ysaen, the Figures Most used info is interesting just to see how far ahead V3 is compared to other figures, 22% compared to Michael 3 and Aiko 3 at 13% and 1.9% for Jessi the highest non Daz figure. If you're a merchant and want to find out which figures to create content for then this will tell you. Sadly it may perpetuate the gap as I guess most will go for the most used figures. I think part of the reason that other figures arn't used so much is that there is less content available for them (if that makes sense). I'm a PHP/MySQL developer by trade so if you want a hand getting this lot into a db and displaying the results on a website give me a shout, I'd be very pleased to help. John.

Considering V3 has been around the longest, its not surprising. Although, given her relatively short lifespan thus far, I think A3 is rapidly catching up. And I noticed that 8 people favor Furrette! Woot! Make that 9 ;)

I know she has been around a long time but I hate seeing poor poser figures left out. Just think of all those runtimes with unloved and ignored figures in them. I always root for the underdog :) John

609 is not great turnout, and the problem with any survey where the participants select themselves is that only a certain kind of individual will select in. In this case, being on a retail site, and a tertiary one at that, it selects in those most likely to be early adopters or otherwise seek out new ways to use their money on Poser, so it is no surprise that Poser 6 comes in so high, and in fact it should have come in even higher.

Using industry standard metrics for adoption, Poser 6 probably comes in at around 74 to 82% of the total user base -- which is why the preference question is so important. Knowing what people prefer to use is more important, and I'm fairly satisified given the wide coverage of the survey (pretty much every major poser site featured it to some extent or other) that it's an excellent pull on the current userbase. Using only the estimated base of community specific active users, 609 respones is more than significant enough a number to be able to apply extrapolated data across the market with reasonable certainty. IT is indeed a great turnout in terms of polling. I'm rather excited now that I've acquired some exctremely effieicnt polling software for the follow up in June -- it will be muchmore smooth an operation with a more immediate and readily accessible dataset that when compared to this one will really rock as a decent comparative. Hell, it's soo good the frequency might even be able to be increased...

The survay may be skewed for other reasons. I tried three times to submit the survey. Each time I got an error and lost what I had filled in. Discouraged I gave up.

I wouldn't mind taking the poll in June. I didn't even know this poll existed until the results were posted. :P And I browse forums every day (which ones? You'll have to wait til June) I agree with the unloved & unused statement, though. As one of the early advocates of Sara (v1), I lobbied hard for more support. And sadly, she sits alone in her runtime folder now, rarely seeing the lights & cameras. Of course, Furrette dominates my images now - and she's a bit of an under... dog ;)

...interesting...but what you're getting, among other things, are stats generated from people who will do on line surveys. A lot of us won't.

I'm aware of that Tunesy. And that's factored into the survey. It's factored into any survey that is meant to cover a particular group. Although I'm ill equipped to explain it in detail, basically, each person's answer is effectively good for 3 or more people depending on several other factors (such as the population sample being looked at). By compiling several statistical results over a space of time, you slowly increase the accuracy level of the survey and require a smaller sample of the population in order to achieve an accurate portrayal. So not filling out a survey doesn't skew the results -- it's only the completed ones that actually skew it. From there, algorithms based on comparative areas or developed specifically to that grouping can be used, increasing the overall accuracy. For my personal analysis, I use a series developed from Software upgrade survey's done between 1991 and 2003. Once again -- it comes down to whether or not ya trust the source. If you don't trust the data, then ignore it. If you do, then use it. I trust the data. Its good stuff, and very useful, and while I'm surprised at a lot of the things that I'm pulling out of it (for example, the age-spending-fave figures info) it also still fits in with what I've observed within the community overall. but, above and beyond everything else: this is fun. I enjoy it -- all the business stuff aside, it's really cool jsut to see the info :D Gives us a better picture of the sort of community we are -- and it's a grand one :D

Very interesting, thanks. I'm a sucker for stats :-)

retarded, cant even work with these files without some further information, the data makes no sence Sorry to hear about your disability, but it explains a lot, I must say. bonni

Ahem... Please discuss the DATA or the survey itself. MachineClaw - please do not be so rude. ynsaen created this survey on her own time, at her own expense, for the benefit of the community as a whole. The fact that you are unfamiliar with raw data usage is not her fault. And being so impolite is hardly the way to ask for the results to be translated into something with which you might be familiar. Manners cost nothing. Thank you. Karen Poser Mod

Very interesting data set. Think I'll load those CSVs into a database myself too. I wonder why you published the IP address data I can see the need of a key, but I think it would have been better to hash the IP addresses. Privacy and Internet are touchy items these days, with good reason. By the way, what database format will you publish in?

Actually, I didn't set up the IP's as the key, but it was my error that allowed such to happen (I didn't say to generate a new key for the results, which was my bad). Fortunately, the next survey won't be limited in such a manner (which is nice -- I'm loving this new software and am considering doing the next survey in April). The database format I'll be releasing it in once I have access to my main system will be access format for this one. After this, the survey data will be released in MySQL formatted CSV, since this new system can export such. Probably would help if I had a better grip on database stuff, but it requires a bit of a different kind of insanity...

Yansen - I know we disagree at times :) But can I make a sugestion for your next survey. [Which could be usefull if the data was sorted or graphed] A suggestion that might get more folks to take part. Lose the salary and age questions. That kinda info is not something I like to share with the outside world. I could've been untruthfull about that, but that skews the validity of the data instantly. If you lie about one part of survey you could be tempted to be not-so-honest with the rest of it. Also it's unlikely to happen, as most IP addresses are dynamic. But using the IP address as a form of 'personal identifer' means you could veer into breeching data protection laws. You need to anonomous it.

Thanks. I'll wait for the Access format. By the way, will that Access DB only contain the data, or will it also contain queries for the different statistical operations you applied? If it's only the data, I'll import them in Access myself, no problem there.

"Lose the salary and age questions." Follow up survey has them as optional. I'm very disinclined to use the ip data (saved automatically as part of the system I used previously) for a key, but I don't know enough about how to alter it or set it up differently, and I had already delayed the results three times, so finally caved in order to not appear to have been "holding back". Despite the occasional attempt at inferring commercial/sinister/whatever motives to this (and there's valid reasons for all such inferences), I'm really more interested in getting the information out and available to anyone who wants to use it. If I get two more requests, I will pull the survey data again for a short time and find a way to rekey it all. Thanks :D ********* The access db will only have the data, and, to be frank, the reason it's not released as of yet is that I can't get into my main system to set it up, and the computer I'm using doesn't have enough power/space to load office onto. Basically, I'm stuck -- I'm really not personally able at this time to do much at all with this stuff. Sorta frustrating, lol. I've been using openoffice.org to set the data up in spreadsheet format myself, and while it's not bad at all, it's slow going as I'm learning that the methods I'm used to using to generate charts and graphs are not the same, lol. That said, I will note that Calc does a pretty decent job for free spreadsheet!

"inferring commercial/sinister/whatever" No inference intended on that. I can see your intentions are good.

What Office version do you use (when your main system is up again, I mean)? Loading those data into Access and replacing the IP numbers by something else is something I can do in less than an hour, I could send the Access db file back to you. I use Office XP UK.

XP would work -- I actualy use 2003 myself. That would be great, actually :D

Thanks for this, ynsaen :D It's great information :D

If you want some help setting up the database for your next survey, give me a holler. Database design is one of my specialties (yep, that other kind of insanity...)

elizabyte...Sorry to hear about your disability, but it explains a lot, I must say. ROTFLBO !!!!...yes, it does explain a lot, doesn't it! ;P hehehehehehehehehehe

ynsaen, "available to anyone who wants to use it" - e.g. spammers and other "do-gooders?" :-O Will be interesting to see if anyone EVER completes one of your surveys again. You promised anonymity, and yet included IP addresses, plus personal information. There was an implied contract of trust from participants, no? Also the US does not have a Safe Habour agreement with the UK or Europe. So all entries with UK/Euro IP addresses contravene their Data Privacy Act. Can't remember the penalties off hand... IMHO you should withdraw this database from R'osity ASAP - if only to protect Renderosity from TPTB.

Penalty isn't applicable. I had already investigated that. The survey is still anonymous. However, I did say if I had three notes I would remove the data and correct that issue. I have three.

ok, due to some people having issues, I've pulled the files off the server until such time as I can correct a problem with them. As this requires me to have my main system up and running, I'm unable to give a date for their further availability at this time. My apologies for the inconvenience.

Well I'll just say thanx for taking the time and effort to have a stab at doing what is an incredibly complex issue. here's to you Ynsaen, don't let the detractors put you off!

Please don't take this the wrong way, but 609 reponses is a pretty sad turnout for a "community survey". After all, the top of this page says "1712 surfers currently online" (and I've seen the number as high as 2400). This means there are 3 to 4 times as many people online right now than took the survey. 609 responses is an especially bad turnout when Renderosity alone claims to have over 250,000 members. So, even though your survey was a very good idea, I don't think it's completely reprsentative of the Poser "community". For example, just because a majority of the 609 respondents said V3 was the most-used figure, you can not say that all 250,000+ people in the Poser community think this is the most-used figure. On the other hand, your survey can make definite statements such as "Of the people who responded, X% prefer using P6." --John

You know when anyone does "market research" which this survey is a form of, they only ever take a "sample size" and a sample size of 609 out of 200,000 isn't that bad a sample size really. Not aimed at anyone ;-))

Well, it might look that way, but that's incorrect. There's a critical response point for polling. That response point is 600. Statistically speaking, a response point of 600 gives you a significance of 1000, meaning that it can be applied, with reasonable certitude to a base of 6 million people. 600 responses is the minimum required by most major polling houses, and in general they cull duplicates from there. Its a funky area, statistical polling -- but yes, John, I can say that the results apply within a roughly 8% variable to the entire community. Can I say what an individual will think? No. THere's no means of doing that. Can I say what the community as a whole will think? Yeah. Human group dynamics does follow known patterns that can be quantified. Your argument seems valid, but it isn't. And I'm not able to effectively refute it without getting into a class on sociological statistics. You can, however, google it out to your hearts content :)

RE:If you're a merchant and want to find out which figures to create content for then this will tell you. Sadly it may perpetuate the gap as I guess most will go for the most used figures. I think part of the reason that other figures arn't used so much is that there is less content available for them (if that makes sense). Looks like Poser user want to use other figures as well. [Quote] There is a HUGE demand for Male clothing for all the figures.[unquote] V3 doesn't enter into if for me. I no longer have Daz UnimeshMil Figures in my P6 Runtime. I much prefer Jessi, James, Miki, Koji, TY2. Cheers DR

Thanks to svdl, the data is once again available. It is now, however, in access 2003 format, and will require that program in order to read the results. The concerns expressed earier have been addressed.

Lose the salary and age questions.

I suspect a lot of people lie about those anyway. Freakonomics has a section about online dating sites, and wonders how so many users can be making $200,000 a year or more, when only 1% of Internet users make that much. ;-)

*After all, the top of this page says "1712 surfers currently online" (and I've seen the number as high as 2400). This means there are 3 to 4 times as many people online right now than took the survey.

Yeah, but how many of those are Poser users? That's the number for the whole site, not just the Poser forum.

I also get the feeling there are a lot of Rosity members who only come here to look at nekkid boobies in the gallery, and are not interesting in using Poser or buying Poser figures or content.

Just to clarify, yet again: THere is no IP information available in the files now. IP information is nt an effective means of establishing identity (if it was, there'd be a lot fewer spammers).

welp, one less point I'd have to make, randym77..thanks..;) Maybe a double-blind survey..;)
If you can export in excel format, comma-delimited, or even text (more steps, but I do it with server logs a lot to squeeze out data. Just replace commas with odd characters like |, import to something like Excel [which open office has an equivalent to], and break into fields by using the special character., etc).
Queries aren't that hard to do, once you know how. you can use a query to strip personalize info out, and maybe export that. Just a thought..

Ynsaen Re:There are 609 effective responses in the final break, and the final accuracy level is approximately +/- 7.3412%. Question: If I see 916 records for User Software in the Survey, How did you arrive at 609? Which ones did you throw out? Cheers DR

To me that would indicate that 609 people responded, but some use more than one software package.

I have done customer satisfaction surveys for the company I work for, last year "my best for return reply" I sent out 1250 surveys and received 324 response those who used the info was ecstatic and congradulated me on a great response so if you got over 600 then that is very very good. besides my survey was sent to those who we told it was coming and gave a prize for responding and still got a low return "IMHO" so congrats, even though it is not "scientific" you can still guage the poser community with this type of info

The Access file in the link in post #1 is in Access 2000 format. The ZIP file also contains the CSV files.

Dang ... wanted to see the survey results but can't use the Access stuff .. Oh well

"I sent out 1250 surveys and received 324 response" In this case, it was not a self selective group like an online poll, but a random sample of your customers, some of which responded. A phone poll works best as it makes it harder for them to select out. You method was second best, and the worst possible data comes out of any method in which the respondants select themselves. Such data is considered meaningless in the political research industry, and I suspect the same holds in marketing pollsters.

...the worst possible data comes out of any method in which the respondants select themselves. but all respondants select themselves. wether they choose to click the website, or mail the form in or talk to the person on the phone. unless your observing from afar then the data is from those who choose to respond. In this case i would think it is the best data avaible. Its not like we can ask E-F to hand over the addresses or phone numbers of people who use poser nor could we get the e-mail addresses from renderosity or DAZ or the like.

On sample size: 609 actually is not bad. In political research you could accurately predict an election off of that. PROVIDED... that your selection means was valid. The utility of a poll is 99% an issue of the means that was used to select respondants, and 1% the questions asked. * Your method must be as random as possible, and must avoid as much as possible any level of self selection in or out by the pollees. On the question of incomes. I find that to be a very valid question. This is a market survey attempt after all - and you need to know what income bracket you should be targeting. If you did a valid random sample survey, and it came back with a majority of users ina low income bracket, you could state that prices on places like renderosity might be too high. By contrast if it came back with a high income bracket, the reverse might be true. But these of course with have to be contrasted against the luxury spending budgets of the respondants and whether they use the product as a hobby or in professional needs. And etc... the market people around here could pick those kinds of questions better than me... I could tell you how to structure a political poll in the basic, given my education, and how to set it up for accuracy (which applies to both types). But I'm guessing for questions on a market poll. That said, it seems a sound guess than income is important.

In this case i would think it is the best data avaible If the 'best data available' is still a meaningless sample, it is still not valid data. Any plan made based upon it is as likely to be misguided as well guided. The data is of no practical utility when it is entirely based on a self-selected sample. There are all sorts of means to improve the sample in a poll, and quite a lot of literature on how to do this. My own education began with: A Practical Guide for Policy Analysis: The Eightfold Path to More Effective Problem Solving (Paperback) Public policy analysis: An introduction (Paperback) Which deal in political research. There is like material available for commercial research. Likely more material available as they have better funding... :) the general gist of it all comes down to - you need to find a means to randimize your sample as much as possible, and to prevent self-selection either in or out as much as possible. While a pure random is impossible, it is not impossible nor even unduely hard to get a fairly random sample. This poll is 'fun' for anectdotal evidence of the online community's most proactive members, but that is all it should be seen as.

based on the methods avaiable how could one obtain a meaningfull sample?