APPENDIX ADEQUACY OF DATA 



665 



Some Ownerships Difficult To 

 Classify 



The possibility of liuman mistakes and faulty 

 judgment affect the reliability of the ownership 

 data just as they affect other data. Training, 

 close supervision, and critical review of results 

 helped to keep such errors to a minimum. How- 

 ever, ownership data are especially subject to two 

 kinds of nonsampling error which are difficult to 

 correct. These are errors of reporting and errors 

 of classification. Reporting errors may arise 

 where public or private records and reports are 

 used in lieu of direct measurement to Forest 

 Survey standards. Classification errors are of 

 particular importance in all private ownership 

 surveys because some farmers operate sawmills, 

 for example, and some forest industrial firms 

 manufacture both lumber and pulp. The possi- 

 bility of misclassification has been minimized as 

 much as possible by using standard definitions 

 and by training enumerators to recognize mar- 

 ginal cases. Nevertheless, the possibility of such 

 error does exist and there is no ready way of 

 measuring it. 



FOREST TREE PLANTING 



The most reliable planting data are those which 

 describe past accomplishments. Since 1926, State 

 foresters have reported — and the Forest Service 

 has compiled — areas planted annually. State by 

 State. There are also fragmentary but reliable 

 statistics available for many years before 1926. 

 In using all of these planting records, judgment 

 enters in only when converting from area planted 

 to area of acceptable plantations. Since an ac- 

 ceptable plantation is defined in terms of number 

 of trees per acre at the end of the fifth year after 

 planting, data for plantations older or younger 

 than 5 years could not be used without allowing 

 for differences in plantation age. The plantablc 

 area data are believed to be adequate, but their 

 reliability covers a wide range. This is because 

 their preparation varied from State to State de- 

 pending upon the availability of local information, 

 and because a considerable degree of personal 

 judgment was usually necessary. 



In general, the estimates of area of acceptable 

 plantations and plan table area were prepared 

 jointly by State foresters and local Forest Service 

 planting specialists. No special surveys were un- 

 dertaken, but full use was made of existing data 

 on stocking of commercial forest land such as the 

 Forest Sui'vey provides, results of planting surveys 

 in some States, public forest records, and similar 

 sources of information. 



Comparability of the planting chita is strongly 

 affected by the local interpretation given to the 

 standard definitions and concepts as explained in 



the section Forest Tree Planting. For example, 

 plantable area is nonstocked or poorly stocked 

 forest land or nonforest land on which, judged by 

 1952 conditions, (a) the establishment of forest 

 tree cover is desirable and practical, and (b) 

 forest tree regeneration will not occur naturally 

 within a reasonable period of time. In each 

 region, attempts were made to insure uniform 

 interpretation of this definition. For example, 

 "reasonable time" was taken to mean 5 years in 

 poorly stocked seedling and sapling areas in the 

 eastern forest types and in coastal conifer types 

 in the West, and 10 years in interior types in the 

 West. The data apply to virtually all of the 

 nonstocked forest land and also to certain areas 

 of seedling or sapling stands that were slightly in 

 excess of 10-percent stocked and where local 

 experience and judgment indicated that planting 

 was practicable. The nonforest land included in 

 plantable area generally pertains to former tim- 

 berland diverted to cropland, but which now lies 

 idle and no longer is used for sucli purpose; non- 

 forest land in use as cropland was not included.^ 



TIMBER RESOURCES OF NORTH 

 AMERICA AND THE WORLD 



The data presented in the section on timber re- 

 sources of North America and the world came 

 from many sources, and many adjustments had 

 to be made to place them all on a common base. 

 No evaluation of these foreign statistics can be 

 offered since none is given in the references con- 

 sulted. The point to be made is that forest in- 

 ventories have never been made in most of the 

 countries of the world outside of North America 

 and parts of Europe. The data are indicative and 

 no great reliability should be attributed to most 

 of them.. 



The estimates for Canada were taken mainl}- 

 from reports of the Canadian Department of 

 Northern Affairs and National Resources. An ad- 

 vance draft of the statement on Canada was re- 

 viewed in the Forestry Brancli of that Depart- 

 ment. The Canadian estimates are believed to 

 be more reliable than the estimates for most other 

 countries. 



The estimates for Mexico are based on frag- 

 mentary data that were brought together from 

 various sources. Major reliance was placed on 

 "Informe al Gobierno de Mexico sobre Silvicul- 

 tura," a report by D. T. Griffiths which was pub- 

 lished in 1954 by the Food and Agriculture 

 Organization of the United Nations. 



The major source of the data given for the rest 

 of the world was "World Forest Resources," also 

 published by the Food and Agriculture Organiza- 

 tion and released in 1955. 



' Cropland which might be planted under various public 

 programs subsequent to 1952 — the "Soil Bank," for ex- 

 ample — is not included. 



4iJil2!»G O — 58- 



-44 



