Sent by Marilyn Matty on 28 May 2002 00:12
on 5/26/02 11:04 AM, Tuttle Grey at [EMAIL-REMOVED] wrote:
>> From: Marilyn Matty [EMAIL-REMOVED]>
>> To: Chris Kaminski [EMAIL-REMOVED]>, CSS-Discuss
>> [EMAIL-REMOVED]>
>> Subject: Re: [css-d] A question of browsers...
>> Date: Sun, 26 May 2002 06:49:10 -0400
>
>>> StatMarket and TheCounter both get their numbers from sites using
>
>> And neither source states that their data or methodology is peer
>
> Oh good lord, "peer reviewed"??? We're not talking about medical research
> here. Market research and good pools done for product companies are also not
> peer reviewed, nonetheless that doesn't eliminate their validity. Enough
> about that though.
I answered question about whether StatMart and TheCounter provided a valid
analysis of the web in general. Any study that makes claims to represent
such a large segment of the population would need to have its methodology
and results peer reviewed.
>
> Naturally StatMarket and TheCounter are going to be skewed by whatever sites
> they are monitoring. That would be true of *any* service. It seems to me
> that, in the absence of any better stats, these two are still worth
> reviewing. Sure they have had lag times before they correctly identify new
> browsers, but that's temporary.
Since they are claiming their information reflects current usage statistics,
and they cannot fairly make this assumption based on flawed methodology and
faulty sampling.
>
> Their stats do seem reasonable when compared to my site logs. The numbers
> are not identical, but it seems to me that they do offer a useful indication
> of how many users are using which browsers, resolutions, etc. We need some
> stats, and unless someone can provide something better, these seem decent.
> And, they are free.
You can scrape up a lot of stuff off the street for free. And for centuries
it was generally accepted that the world was flat. Just because something
"is" doesn't mean it is right, and there are alternate sources of
information for browser stats at no cost.
And I hope you will be as lucky if you go to Las Vegas as you were in
finding a reasonable correlation between your stats and those of these
studies.
>
> CodeBitch wrote an article demonstrating how her own site stats were quite
> different. Of course they were. The sites for *any* individual site will be
> different, as that is a much smaller sample. Those sites are by definition
> much more skewed, as the sample is much narrower.
CodeBitch's article was excellent, I think. And she correctly inferred that
if you a benchmark for browser stats for a low volume site, the best
alternative to spending big bucks is server logs. It's the best way to
determine how deep into CSS you can build a site.
>
> If you have good stats on *your own* site, then clearly that is great
> information. The visitors you most want to know about are your own visitors.
> In the absence of that, or as a supplement to it, these large stats sites
> seem useful even if the specific numbers might actually vary a few points
> from what they report.
The error margin for a valid survey is +/- 3%, and I think these two studies
deviate far from it.
>
> If someone can point out a free stats service that has better stats, based
> on broader sampling of sites and visitors, then that will be good for all of
> us. Until then we must use what we have, and from what I've seen and the
> discussions I've read, these two big services still offer food for thought
> and helpful information.
There's always a Magic 8 ball; I saw them really cheap at Costco a few weeks
ago.
Marilyn
