About statistics... If you have a universe of 50.000 "users" you need to measure 2.381 to have a valida data with an error of a 1%.
the formula is "n=NPQ/[e2(N-1) +PQ]" To have some bad "users" in our universe (for example - people who goes to the bathroom while visiting your site), doesn't mean the sample is bad... that people will be in our 1% or error. Please, one bad user doesn't abolish the sample. In the other hand, I did a study, as you comment, about people coming from different partners sites and yes... people present different behaviours. Some sites give you links to the home, other to specific products, others to the registration page... You need to understand all about your web site to understand what the log is telling to you. Cesar. On Mon, 18 Nov 2002 15:00:32 -0600 James Riemermann <[EMAIL PROTECTED]> wrote: > That would be quite an anomaly, if hundreds of > visitors from one site were > systematically 3-4 times as likely to go to the > toilet while browsing your site as > hundreds of visitors from another site. In > fact, it would be a statistical miracle, > unless one of the referring sites were selling > laxatives with instantaneous home > delivery. > > I'm not saying that this comparison tells you > that the average user of one site is > reading your site for 200 seconds, the other > for 60 seconds. That would be an > enormous assumption, probably wrong, because > you *don't* know what a user is doing > between server requests. But you do know there > is a large discrepancy between the > average session lengths from the two referring > sites, and it is only a modest > assumption that, with a sample of hundreds of > users, the higher numbers represent a > genuine and substantially higher amount of time > viewing your site. As I said, you > need to apply a little common sense when > interpreting these numbers. > > james > > >>> [EMAIL PROTECTED] 11/15/02 01:44PM >>> > Well yes, it could be not meaningless, but then > you still dont > know the reason why there is that difference in > session time between the > two sites. > Maybe the visitors who visit website_B with an > average session > length of 200 have an urge, while viewing the > website to go more to the > toilet while watching a page on the website, > then visitors who visit > website_A with an averige of 60 in length of > session time. > > But then I cant image you are interested in the > average time > a website visitor is spending time on the > toiled while he/she has a > page open on his/her computer.!!! > > So it is still bull..... > > My two cents, > > +------------------------------------------------------------------------ > | This is the analog-help mailing list. To > unsubscribe from this > | mailing list, go to > | > http://lists.isite.net/listgate/analog-help/unsubscribe.html > | > | List archives are available at > | > http://www.mail-archive.com/analog-help@lists.isite.net/ > | > http://lists.isite.net/listgate/analog-help/archives/ > | > http://www.tallylist.com/archives/index.cfm/mlist.7 > +------------------------------------------------------------------------ > +------------------------------------------------------------------------ | This is the analog-help mailing list. To unsubscribe from this | mailing list, go to | http://lists.isite.net/listgate/analog-help/unsubscribe.html | | List archives are available at | http://www.mail-archive.com/analog-help@lists.isite.net/ | http://lists.isite.net/listgate/analog-help/archives/ | http://www.tallylist.com/archives/index.cfm/mlist.7 +------------------------------------------------------------------------
