<br><br><div><span class="gmail_quote">On 10/12/06, <b class="gmail_sendername">Doug Henwood</b> <<a href="mailto:dhenwood@panix.com">dhenwood@panix.com</a>> wrote:</span><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<br><br>Your pal Kaplan in Slate made a big deal out of the wide confidence<br>intervals in the old study. Well, check out the CIs in the new one:<br>654,965 (392,979–942,636). So it's somewhere between 400,000 and a<br>million, with the best bet around 650,000. Yeah, that 400k is
<br>reassuring!<br><br>Doug<br><br></blockquote></div><br><br clear="all">Doug or anyone else,<br><br>Are the methods used in the study described in detail on-line anywhere? <br><br>Are criticism of the kinds of methods used in this study on-line?
<br><br>Jerry<br>