On Thu, Mar 21, 2013 at 12:35 PM, Kay Schenk <kay.sch...@gmail.com> wrote: > On Tue, Mar 19, 2013 at 6:25 PM, Kevin Grignon > <kevingrignon...@gmail.com>wrote: > >> KG02 - see comments inline >> >> On Mar 19, 2013, at 6:11 AM, Kay Schenk <kay.sch...@gmail.com> wrote: >> >> > On Sun, Mar 17, 2013 at 6:09 PM, Kevin Grignon < >> kevingrignon...@gmail.com>wrote: >> > >> >> KG01 - see comments inline. >> >> >> >> On Mar 18, 2013, at 1:59 AM, Rob Weir <robw...@apache.org> wrote: >> >> >> >>> On Sun, Mar 17, 2013 at 1:10 PM, RGB ES <rgb.m...@gmail.com> wrote: >> >>>> 2013/3/17 Kay Schenk <kay.sch...@gmail.com> >> >>>> >> >>>>> In light of our recent start on Strategic Planning, is it worthwhile >> to >> >>>>> investigate some sort of survey mechanism? >> >> >> >> KG01 - Graham and I set up a LimeSurvey account, which he hosts. I am >> the >> >> tool admin. >> >> >> >> >> >>>>> I know we've used Google >> >>>>> Moderator, and I see there were mixed opinions on that. >> >> >> >> KG01 - yes, this more a discussion tool >> >> >> >>>>> We really need to >> >>>>> get a better feel of WHAT our end users are doing with AOO I think, >> >> before >> >>>>> we can get into any indepth product planning. >> >> >> >> KG01 - yes, this was our goal when we launched the survey effort last >> >> year. Actually, the survey process is documented in the AOO UX wiki. We >> >> developed the baseline demographic questions which accompany topic >> specific >> >> survey topics. >> >> >> >>>>> >> >>>>> See related thread on using BZ: >> >>>>> >> >>>>> http://markmail.org/message/5a4j74e4oths55rg >> >>>>> >> >>>>> Ideally, we should attempt to use some mechanism that doesn't >> >> explicitly >> >>>>> require an account setup. >> >>>>> >> >>>>> I started thinking about this a few days ago and saw that there is a >> >> MWiki >> >>>>> extension called "Survey". Maybe we could use this. >> >>>>> >> >>>>> Coupled with that, is there some way to setup an automatic "login" >> for >> >> such >> >>>>> a page display, but coupled with CAPTCHA for survey submission >> maybe? >> >>>>> >> >>>> >> >>>> +1 for a survey tool. I remember "LimeSurvey"(1) being discussed on >> this >> >>>> list several month ago, but I don't remember the output of that >> >> discussion. >> >>>> >> >>>> (1) http://www.limesurvey.org/ >> >>>> >> >>> >> >>> Right. Google Moderator is more a brainstorming or "ideation" tool. >> >>> It is not really a survey tool. Something like LimeSurvey is much >> >>> better for surveys. >> >>> >> >>> We have a few options there: >> >>> >> >>> 1) See if we can get it hosted here at Apache, on a VM/BSD jail. >> >>> >> >> KG01 -this approach is more effort, but highly scalable as it supports >> the >> >> periodic succession of hosting and research volunteers. +1 >> >> >> >> >> >>> 2) Have a volunteer host it on their own survey. If we only do >> >>> anonymous surveys and don't collect personally identifying information >> >>> I think this would be low-risk. >> >> >> >> Kg01 -Currently doing this, just need more resources to deploy surveys >> and >> >> take action. >> >>> >> >>> 3) With either of the above options we could assign it a subdomain >> >>> like surveys.openoffice.org >> >>> >> >>> At one point Graham was looking into #2, but that was a while ago. >> >> >> >> KG01 - Graham and Kevin and Graham. See notes above. >> >> >> > >> > Kevin, does limesurvey require some sort of registration by participants? >> > >> > It looks like it does. >> > >> >> KG02 -Hhmmmm good question. I hope not, lets verify. >> > > It would be great if someone COULD verify the needed logon business. > > OK, more questions and an ultimate goal. I have not kept up with marketing > on any of this, so here goes... > > * Does marketing have results of previous surveys actually published > anywhere? > > I took a brief look at the Marketing area on the web server just now and > could't find anything. It would be great if we COULD publish this > information someplace, at least for a time -- a year or so. The results > might be valuable for planning purposes. > > * Can we put together a survey to find out more about what users are doing > with AOO? > > We have BZ for bug reporting but I think it would be nice to actually find > out what folks are doing with AOO -- who they are, what modules they use, > and for what if they care to share that information. My personal interest > is with Base. What are people doing with it. > > * Ideally, we should use some survery/mechanism site that does NOT require > a login -- some simple human verification like CAPTCHA is fine. The last > thing we want is to inhibit people from participating. >
According to this page they offer both anonymous and non-anonymous surveys: http://www.limesurvey.org/en/about-limesurvey/features > Thoughts? > It depends on the purpose of the survey and what you want to do with the results. In general, an open survey, where anyone can respond will not give the best results. It can get a lot of responses. It can be used as a "push poll" to influence options. But the main problem is that the respondents are self-selected and when that happens you risk that they are not representative of the entire group you are surveying. It might be that those who responded to an advertisement of the poll on Facebook or the mailing list, might younger, or more likely to be a native English speaker, or biased in any number of other ways. Ideally -- and we might not always be able to do the idea -- we start with a list of representative users and make a random selection and invite the random subset to fill out the survey. If someone doesn't respond, then follow up with a reminder. And a 2nd, personal reminder. You try to avoid a bias caused by non-responders being different from responders in some important way, e.g., they are busy, or they don't have broadband, or whatever. If you get an actual representative sample then you can answer almost any question with a survey of only 400 people. But if the respondents are not representative of all users then you can have 400 thousand responses and have the results be meaningless. That is the challenge of surveying -- getting the good random sample to start with. Something built into the product itself would be idea, since that has the potential of reaching all users equally. Regards, -Rob > > >> >> > >> >> >> >>> >> >>> -Rob >> >>> >> >>>> Regards >> >>>> Ricardo >> >>>> >> >>>> >> >>>>> >> >>>>> -- >> >>>>> >> >>>>> >> >> >> ---------------------------------------------------------------------------------------- >> >>>>> MzK >> >>>>> >> >>>>> "Achieving happiness requires the right combination of Zen and Zin." >> >>>>> >> >>> >> >>> --------------------------------------------------------------------- >> >>> To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org >> >>> For additional commands, e-mail: dev-h...@openoffice.apache.org >> >>> >> >> >> >> --------------------------------------------------------------------- >> >> To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org >> >> For additional commands, e-mail: dev-h...@openoffice.apache.org >> >> >> >> >> > >> > >> > -- >> > >> ---------------------------------------------------------------------------------------- >> > MzK >> > >> > "Achieving happiness requires the right combination of Zen and Zin." >> >> --------------------------------------------------------------------- >> To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org >> For additional commands, e-mail: dev-h...@openoffice.apache.org >> >> > > > -- > ---------------------------------------------------------------------------------------- > MzK > > "Achieving happiness requires the right combination of Zen and Zin." --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org For additional commands, e-mail: dev-h...@openoffice.apache.org
