APPENDIX — ADEQUACY OF DATA 



657 



Since the bulk of timber products output was 

 estimated with a small sampling error, a relatively 

 large sampling error could therefore be tolerated 

 for products other than saw logs, pulpwood, and 

 veneer logs and bolts without exceeding the 10 

 percent sampling accuracy goal. Thus, it appears 

 that the sampling error of the total 1952 estimate 

 of timber products output in the United States 

 and Coastal Alaska is about 2.0 percent, and the 

 average of 6.5 percent per billion cubic feet is well 

 within the accuracy goal. By products, the 

 sampling errors are as follows: 



Sampling Sampling 



error of error per 



total hillion 



estimate cubic feet 



(.percent) (percent) 



Saw logs 2.5 6.3 



Pulpwood 



Veneer logs and bolts 6.4 1.1 



Other products 5. 6 9. 



AU products 2.0 6.5 



For saw logs and veneer logs and bolts, the 

 sampling error per billion board-feet averages 15.8 



percent — 16.1 percent for saw logs alone and 2.8 

 percent for veneer logs and bolts. 



The approximate sampling errors of other break- 

 downs of 1952 timber products output data can be 

 read from figure 2. For example, the total hard- 

 wood output in the West Gulf Region is 0.5 billion 

 cubic feet for which the figure shows a sampling 

 error of slightly more than 9 percent. For regions 

 or States listed in the tabulation above, figure 1 

 also can be used for approximating the sampling 

 error of breakdowns in the same manner as for 

 area, volume, and gi'owth data. 



NONSAMPLING ERRORS ArE POSSIBLE 



The timber utilization data are subject to errors 

 other than those which arise from sampling. For 

 example, mistakes may have been made because 

 lists of timber products consumers or producers 

 were incomplete, or because production or con- 

 sumption reports obtained by mail and field can- 

 vass were in error. There may have been mistakes 

 in selecting and applying converting factors, in 



100 

















































































80 















































60 

































Sw 









































^ 

































40 































S^ 



































30 

























V, 



•s 



S 



^ 



























_ 20 



1 — 



^ 



























S 





V, 



s 



i 



R BC 



lARO- 



FOOl 



■ u 



JITS 











LXJ 



- 10 



X 



v 























S 













^^ 

















C5 fi 











•^w. 











>s 















^ 



^ 











N, 



S 



v^ 











0£. 



^ 6 













^ 















FO 



R CL 



Bl( 



:-F 







ITS^ 



•v 











^ 











o 



OOT UN 



s 









El 4 























Xw 

























^ 



■v 



s, 















^ 3 























k^ 































s 



^^ 







^ 



^ 



2 





























^ 



•s 



V 













































1 





































.07 .1 .2 .3 .4 .5 .6 .8 1 2 3 4 5 6 8 10 20 30 4 



TIMBER PRODUCTS OUTPUT IN BILLIONS-BOARD-FEET OR CUBIC FEET 







Figure 2. — Sampling error of timber products output in percent. 



