APPLYING THE CHEGE SCALE CORRECTION EACTOR 



The usual and simplest method of computing the correction factor 

 from the check scale data is to divide the checker's net voluiTxe by 

 the estimator's net volume for the same trees. The total estimated 

 volume obtained by the first sample or by a complete tally is then 

 multiplied by the correction factor to give the final corrected net 

 volume. The method is based upon the assumptions that (1) a line 

 fitted to the plotted points of check scaled volume over estimated 

 volume passes through the zero origin, (2) that the best fitting 

 regression for the plotted points is a straight line, and (3) that 

 the variation about the straight line is proportionate to the volume. 

 In most cases, these assuraptions are sufficiently correct for 

 practical purposes. As has been pointed out previously, if the 

 relationship does not meet these assumptions, a separate correction 

 factor for size strata should be usedo 



In no case should the mean of the check scaled volumes be multiplied 

 by the number of trees in the sale to determine the total volume. 

 Multiplying the cheek scale mean volume by the total number of trees 

 ignores the information of the first sample or complete tally and 

 employs data from only a relatively few trees o It is not accurate, 

 A check scale on a few representative trees can give a reliable cor- 

 rection factor v/ithout necessarily giving a reliable mean volume. 



Contracts for. timber sales on which tree volumes are to be determined 

 by sam.pling must have strict rules concerning adjustments. A practice 

 of adjusting the correction factor vjhen a sale volume is c^uestioned 

 by the operator could result in a consistent loss to the seller, 

 because underruns vjould bring complaints and overruns probably would 

 not be mentioned. 



TEE POSSIBLE SAVING BY SAMPLING 



Sampling can substantially reduce the cost of tree measurem.ent . 

 Although the costs have not been determined by accurate studies, they 

 can be roughly estimated. Discussion with many forest officers and 

 personal experience have led to the belief that marking and measur- 

 ing 75 to 100 trees per man-day is an average accomplishraent on tree 

 measured sales. Under winter conditions daily production may be 

 lov;er, but 100 trees per man-day is probably a safe average. The 

 same sources of information give reason to believe that an average of 

 250 trees per man-day can be marked, v;ithout measuring. If 4p20 per 

 day is a fair estimate of the cost of one man-day, counting all ex- 

 penses that are associated with the job, then 20 cents a tree is the 

 cost of marking and measuring. As marking alone costs 8 cents per 

 tree, the cost of measuring a sample tree is about 12 cents. 



-25- 



