SURVEY OF FOUR TOWNSHIPS IN SOUTHERN NEW HAMPSHIRE. 



vailing types of farming, and the profitableness of these types for 

 that section; also to learn the distribution of capital, income, and 

 expenses, so as to gain a correct view of the agriculture as it now 

 exists. The plan of the work was to visit personally each farmer 

 within a certain area and obtain from him the information desired. 

 In tliis way fairly accurate data were collected and these were checked 

 up by careful inspection on the part of the person making the survey. 



ACCURACY OF THE RESULTS OBTAINED. 



The information obtained has been carefully tabulated and aver- 

 aged so that the re- 

 sults as given may be 

 considered fairly ac- 

 curate. This accu- 

 racy depends on the 

 law of averages. For 

 instance, the individ- 

 ual farmer in giving 

 the amount of his ex- 

 penses for the year 

 might make a consid- 

 erable error, but he is 

 just as apt to overes- 

 timate as to under- 

 estimate. According 

 to the law of aver- 

 ages, therefore, when 

 a considerable num- 

 ber of farms are taken 

 into consideration 

 the amount of the 

 overestimate will 

 about equal the 

 amount of underesti- 

 mate, so that the av- 

 erage is for practical 

 purposes fairly accu- 

 rate. 



To explain tliis a httle more fully, suppose that a hundred men 

 who know something of the weight of horses estimate the weight of a 

 given horse. Some of them will estimate too much and others too 

 little. But if the men have a fair idea of the weight of horses the 

 average of all their estimates will be very close to the correct weight 

 of the horse. The figures here given are in each case only averages 

 of a considerable number of farms. 



[Cir. 75] 



Fig. 1.— Diagram showing the decrease in improved and the increase 

 in unimproved farm land in New Hampshire from 1850 to 1900. 

 (U. S. Census.) 



