On Wed, 20 Aug 1997 08:59:04 -0500, Sharyn Horowitz <[log in to unmask]> wrote: >I try with a few people (ie 3-4), fix what's not working, and repeat. I >keep going until visitors seem to get it. It's not scientific, but it >works, and it's a lot more satisfying than 100 trials of an exhibit that >doesn't work. > >>Hi--For those of you who have done formative exhibit evaluations, what >>is considered a sufficient sample when observing visitors interacting >>with a small exhibit with only one "interactive" element (a touch box)? >>100? 200? Any other thoughts on formative evaluation appreciated (and >>yes, we have the book "Try It" which is on this very topic). Thanks for >>any feedback! >> >>Stacey Otte >>Collections Manager >>Catalina Island Museum >>[log in to unmask] > >Sharyn Horowitz, The Health Museum of Cleveland >[log in to unmask] >8911 Euclid Avenue >Cleveland, OH 44106 >(216) 231-5010 >fax 216-231-5129 Sharon's right on the button with her response. If it takes about four iterations to "get it right", that's a total sample per interactive of 16-20. If it's not working after 4 tries, seriously rethink, or give up, on that one. If you know you have different target audiences for the material, one, you're a genius :-), and, two, you may have to test some more for each audience. The trick is to correctly set the scope of what you're testing. A big, complicated interactive with many intuitive leaps will take more testing, and maybe should be broken down into manageable chunks. I would quibble, in the nicest possible way, with Sharon's, "It's not scientific...." It is. Sample size is only a tiny part of "science." Doug Note new email address: ************************************************************* Doug Hoy Evaluation National Museum of [log in to unmask] & Science & Technology (613)998-6863v Research P.O.Box 9724, Station T (613)990-3654f Ottawa K1G 5A3 CANADA **Opinions expressed are not necessarily shared by the NMSTC* *************************************************************