30 BULLETIN 1382, U. S. DEPARTMENT OF AGRICULTURE 
With the percentage of expenditures per cost-consumption unit for 
advancement as the dependent variable, only 16 per cent of the total 
variation could be accounted for by the independent variables used 
in the analysis. Less than three- fourths, 08 per cent, of this 16 per 
cent of the variation which could be accounted for, or 10.9 per cent 
of the total variation, is due to criteria of the ability to pay and 32 
per cent, or 5.1 per cent of the total variation, is credited to factors 
indicative of the desires or demands of the family for the economic 
goods of family living. 
Of the variation which could be accounted for the largest share is 
attributable to net worth of the farmer. This share or percentage 
amounts to about one-fourth of the variation which could be ac- 
counted for with each of the three dependent variables. Interpreted 
in proportions or percentages of the total variation this one-fourth 
amounts to 12 per cent of the total with expenditures per family, 
10.2 per cent of the total with expenditures per cost-consumption 
unit, and 4 per cent of the total with the percentage of expenditures 
per cost-consumption unit for advancement. This does not imply, 
however, that net worth of the farmer is the most significant index 
of the standard of living. About as much of the variation which 
could be accounted for is attributable to size or magnitude of the 
farm business, measured by acres and cost of operation, as is due to 
net worth of the farmer. It is probable that both acres and cost of 
operation are reflected through net worth to the standard of living. 
PRESENTATION OF INFERENCES OR CONCLUSIONS 
Owing to the small number of records from which the data were 
obtained and the wide variations between localities, tenure status, 
and family make-up, inferences or conclusions here presented can 
not be regarded as absolute or final. They are suggestive of further 
study of a larger number of records representing specific regions 
and types of farming. Among the inferences drawn, the following 
are presented as suggestions for further study : 
(1) Economic advancement, reflected in the ability of the farmer 
to pay or to provide, bears a fairly close relation to the standard of 
living. This does not imply that family living is not curtailed in 
many instances in order that the farmer may advance economically, 
that is, increase his productive capital at the expense of the family's 
welfare. Nor does it imply that schooling of all members of the 
family and participation of the family in social activities are not 
reflected in the ability to pay as well as in the demands for the eco- 
nomic goods of family living. 
(2) Both size of family and family-living cycles — that is, periods 
of growth and development — are directly related to the standard of 
living, although apparently to a less degree than is the ability to 
pay. But some of the actual relation between the demands on 
family living, from these causes, and the ability to pay may not be 
accounted for because of the counteracting influence of an increased 
number and age of persons composing the family on the ability to 
pay. 
(3) Schooling of the parents bears less relation to the standard of 
living, in terms of expenditures, than does economic advancement — 
that is, the ability to pay. Allowance has not been made, however, 
for efficiency, due to schooling, in buying and using goods, Fur- 
