On Sun, Mar 14, 2010 at 09:14:06PM -0500, Rex Dieter wrote: > Peter Hutterer wrote: > > > On Sat, Mar 13, 2010 at 09:14:48PM -0700, Orion Poplawski wrote: > >> > >> On Sat, March 13, 2010 4:58 pm, Peter Hutterer wrote: > >> > Isn't there a mere RISK to lose 70-80% of our users if we do _not_ > >> > implement > >> > the changes as well? Especially given the chance that the poll did not > >> > represent a significant user sample? > >> > >> How many users do we need? > > > > sorry, I'm not sure I understand the question. Which user number are you > > referring to? > > I'd venture he meant in response to your "the poll did not represent a > significant user sample" comment. So, how many users are needed to make it > representative? I don't know what the numbers are to make it significant. I'd probably be happy with anything in the range of a couple of hundreds or more. Maybe someone better versed in statistics can chime in here. The tricky problem with this is that the pool the participants are chosen from matters a lot. For example, if you go out on the street during the soccer world championship finals and ask people what they think of soccer, the results are likely to be skewed because most people that are really into soccer will be at home watching the game (this is an analogy only and will present some flaws in direct comparison to the issue at hand). This skewing of results presents itself with any form of survey, whether it be forums, mailing lists, irc, etc. Numbers are easy to get given our user numbers. Chosing a good sample is what is much harder. And don't get me wrong, I don't dispute the results of Adam's survey. My point is that we need to be careful on the interpretation of data and be very conservative - regardless of what the data actually says. Cheers, Peter -- devel mailing list devel@xxxxxxxxxxxxxxxxxxxxxxx https://admin.fedoraproject.org/mailman/listinfo/devel
