From: Jason P. <ja...@jo...> - 2002-02-26 13:49:06
|
Well, if I were using the software for a "survey", per se, then I would agree. However, I'm using the software to build quick survey-like forms that users fill out qualification questions that may change over time. We don't really care about the scientific plausibility of the survey, just to associate the "meta data" that the "survey" collects with users in a separate database. Example Question: "What technology are you intending to purchase within the next 12 months?" [] Technology A [] Technology B [] Technology C Later on, let's say "Technology B" is outdated and needs to be removed. Or a new Technology (D) becomes available and we would like to poll for it. Perhaps this isn't the intended use for the software. ;) I got a similar off-line response from someone else and it has gotten me to think about the whole process... now if I can only show the Boss the way. ;) Thanks, Jason PS: Thanks for building this software - it's really cool! If memory serves me right, on Tuesday 26 February 2002 01:44, James E. Flemer wrote: > So what happens if you activate a survey, some people take > it, and then say you change an option on a dropdown list? > The people who've taken it already have their responses > saved in the database, keyed to the first set of options. > Then some more people take the survey, and their responses > get saved in the database, using the same keys -- but > corresponding to different values. > > What good is a survey, if you don't know what the > respondant filled out!? > > The "recomended" way to handle the situation is the > following... > > First, be more careful. The "test" feature is there so that > you can say, "Hey Boss, this is what our survey is going to > look like, what do you think?". > > Second, if you do Activate a survey then need to change it, > do the following. > > 1) End the first as quickly as possible. > 2) Copy it. > 3) Edit the copy. > 4) Activate the copy, and put it up in place of the > original. > > Yes, this way you lose any responses from before you ended > the first one, but at least your data is consistent for > both (the old and new) copies. > > If you *really* need to edit active surveys, then feel free > to add all the necessary checking to make sure that you do > not muck up key-value pairs ... but don't ask me to do it. > It is not trivial, and I certainly don't have time to do it > (for free). > > -James > > On Tue, 26 Feb 2002, Richard M. Pavonarius wrote: > > I second that motion. I activated a survey a few hours ago and just > > realized I forgot to polish up the confirmation page. Too late now. I'm > > just glad I didn't put something like "Thanks, pinhead." as a temporary > > filler. > > > > Rich > > > > At 15:10 02/02/25 -0500, you wrote: > > >I noticed that I could update values in the survey table to reset the > > > values for title and info, as well as update required in the questions > > > table, all of which would make the small necessary changes that were > > > needed. > > > > > >My manager wants to also be able to add/delete answers to > > > multiple-choice questions - is this possible also through updating the > > > database & SQL? What impact will that have? > > > > > >I added a feature request on Sourceforge to support editing surveys, > > > even after they're Active ... I don't understand why the state of the > > > surveys are so strict or uni-directional. > > > > > >Jason > > > > > >_______________________________________________ > > >phpESP-general mailing list > > >php...@li... > > >https://lists.sourceforge.net/lists/listinfo/phpesp-general > > > > _______________________________________________ > > phpESP-general mailing list > > php...@li... > > https://lists.sourceforge.net/lists/listinfo/phpesp-general |