SI
SI
discoversearch

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor.  We ask that you disable ad blocking while on Silicon Investor in the best interests of our community.  If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.
Gold/Mining/Energy : International Precious Metals (IPMCF)
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  
To: GlobalMarine who wrote (27847)11/20/1997 8:28:00 PM
From: E. Charters  Read Replies (1) of 35569
 
Students T test or chi nova or any stats test that compares variances
of two groups of assays is used to compare assay houses or groups or different labs on pulps or rejects. Because of the huge variability on heads or coarse rejects you would need at least 100 samples to compare results on splits of coarse samples. Fine pulps SHOULD have close comparisons. RE-assay of pulp should come to 95% of first split.

The mean of any split group (or two groups of the same material) will be close even if one lab is horrible. The best method has the smallest variance. That means if the sample to sample variance on a whole group is least then it is likely that random problems in methods also have the least input. Capiche? So if one guy is getting from one ounce to .01 and the other .04 to .50 then the latter is preferred. Now compare the two groups with a T test. The T test tries to determine if they are the same group by comparing variances, and assumes measurement errors crept in. If the T test shows the two groups are 95% probable the same group then the assays from both labs are comparable.

Another way to do it is to combine two assay groups, if the variance changes drastically then we have a problem.

The Variance is got from subtracting the mean or average from each value and squaring it. You then take the average of all these squares.
The square root of that is the standard deviation. Within two standard deviations on each side of the mean should be 66% of all your assays in any one group. Normally.
Report TOU ViolationShare This Post
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  

HomeMailHotSubjectMarksPeopleMarksKeepersSettings
Terms Of UseContact UsCopyright/IP PolicyPrivacy PolicyAbout UsFAQAdvertise on SI
© 2026 Knight Sac Media.  Data provided by Twelve Data, Alpha Vantage, and CityFALCON News