Statistics...

[irenicspace]

I am working this morning to pull together some statistics on web browser use. I found different kinds of warnings about the collection of browser statistics. I like the second one better:

1. Caution: statistics can mislead. Caching distorts raw data; each site attracts different audiences, with different demographics; each survey uses different

Comments

Re: Webtrends

[irenicspace]

I am interested in broad internet statistics, not per site.

PS: I have received no address email yet... ;)

Exactly

[slq]

short reporting periods and small sample sizes exaggerate fluctuations;

This is exactly what I was talking about in my email to you earlier about OTO demographics. If you take a sample of 1-2 years and 100 members from the overal data of 15 years and 3000 members, you can't generalize what you find to the data as a whole.

Re: Exactly

[irenicspace]

I understand your point very well. I also understand Mark Twain's very well too.

3000 members over 15 years might be a good thing, or it might be pathetic. Depends on goals, metrics, direction, etc. Being that the OTO is currently lacking them all, I can see how 3000 members over 15 years might seem like a good thing. However I am comparing the OTO to other organizations in the same time span or less and judge it to be lacking. I guess it is all realtive.