Laserfiche WebLink
Dan <br />From: Mathews, Dan (~[/r ~ <br />Sent: Friday, December 15, 2006 2.44 PM <br />To: Brown, Sandy; Gorham, Kent <br />Subject: Golden Eagle 2006 Production Data <br />Sandy and Kent: <br />Awhile back you asked me to take a look at the subject data, to confirm that the consultant used the proper statistical <br />procedures. I did confirm this at the time; everything looked copacetic. The only question we had was in regard to the fact <br />that the undesirable forb component was not identified, and was included in the data totals. I didn't have time at that point <br />to run the analyses myself, but I was able to do that today (I recently got StatGraphics loaded back on my computer and <br />wanted to try it out; this data seemed like a good candidate). <br />Along with some summary statistics, I ran reverse null t-test comparisons for the Rangeland and AVF areas. In both <br />cases, I subtracted out the Undesirable Forb component, and ran the comparisons on that basis. In both cases, the null <br />hypothesis of equality was rejected; the alternative hypothesis (Reference area mean less than Reclaim area mean) was <br />accepted. This was pretty much a statistical confirmation of what was intuitively obvious from the data. The t-tests <br />demonstrated that the reclaim area means exceed the respective reference area means at the 80% level of statistical <br />confidence, pursuant to Rule 4.15.11(2)(c). If necessary, I would have run the test of reclaim area mean vs (90%) of <br />reference area mean, which the rule allows, but in this case it wasn't necessary (i.e. demonstration was made for 100% of <br />reference area mean, ergo would also be the case for 90% of reference area mean). <br />I'll put a copy of the StatGraphics printouts in the mail. <br />Daniel T. Mathews <br />Colorado Division of Reclamation, Mining and Safety <br />Grand Junction Field Office <br />101 South 3rd, Suite 301 <br />Grand Junction, Colorado 81501 <br />daniel. mathews(o)state.co. us <br />Phone (970)242-5025 <br />FAX (970) 241-1516 <br /> <br />