128 MISC. PUBLICATION 218, U. S. DEPT. OF AGRICULTURE 
TaBLE 41.—Coefficients of dispersion by property classes; New Hampshire and 
orth Carolina } 
State, subdivision, and period covered Forest Farm Other All 
classes 
New Hampshire, 1928: 
Town of Fremont: Pereent | Percent | Percent | Percent 
SSH KG (a) A Fees Mane Sn yg RSE oy Bt AG OE 9 ee LEA Pen) eee ie ae 25 24 18 24 
INOMTESI Gd eb 2 Sees SORE ea Ee ee cece eel wee 37 19 23 34 
OC RT 65 5 2k OS oP se I yl ee ee ree CA ea 38 23 19 28 
Town of Richmond: 
ESIC em Gee: Th ey ee A ta eee Sa OE 26 23 18 25 
INonresid 6nG2= haa es te ey Ua Sole ats Se a 34 12 27 33 
STO Cee es Ee A Set IE CE SNES Wee Ait Sod Ee SOE 32 22 20 30 
North Carolina, 1925-30:2 
Beaufort County aes HOLE SD chao Berek eee A Be Riana 27 29 47 31 
Chatham County 22232 Soe ee ee ee 4 27 46 32 
Macon! County Bene Ae ie Oe ee ee 55 30 32 37 
1 Source of data: Refer to tables 26 and 29. 
3In North Carolina, Forest includes transitional and Farm includes pasture lands. 
It might be expected that within property class groups the varia- 
tions in assessment ratios would be less than in the entire political 
unit of which they are a part. This appears to be generally the case 
in the homogeneous groups. The most homogeneous groups repre- 
sented in tables 40 and 41 are timber in table 40 and farm in both 
tables. In spite of the small size of the timber sample in most 
counties, the timber group shows a lower coefficient than the county 
as a whole in 13 out of the 17 counties in which timber is represented 
by more than six sales. All of the four exceptions are in Oregon. 
The farm group shows a lower coefficient than the county or town 
as a whole in 29 out of the entire 33 counties and towns. The cut-over 
class is not so uniform in character as either farm or timber. It 
includes very low value properties, which tend to have extreme 
ratios as shown by the frequency distributions. This group (table 40) 
has 8 cases out of 27 in which the degree of inequality within the class 
appears to be greater than that of the corresponding county taken as 
a whole. The resort group and the residential and business group as 
segregated in Wisconsin are even more heterogeneous, and the resort 
group is especially unstable in market value. In many counties one 
or both of these classes show greater inequality in assessment than 
the cut-over class. The forest class in New Hampshire and North 
Carolina (table 41) is also one of the most heterogeneous, because it 
was impracticable in these States to segregate the properties with 
old-growth timber from those in different stages of second growth or 
from those which were partly in other than forest use. In this class 
4 out of 5 counties or towns show greater degrees of variation within 
the forest class than within the corresponding political unit as a 
whole. In making the preceding comparisons, it is well to bear in 
mind the possibility that the coefficient of dispersion for the entire 
county or town may be influenced more than it sho d be by the 
dispersion which characterizes a particular class if that class happens 
to be over-represented in the sample. 
In both of the New Hampshire towns there appears to be a tendency 
to assess resident-owned properties with a greater degree of equality 
than nonresident-owned properties, except in the farm category, 
where the reverse is true. The greater inequality in assessment of 
forest properties owned by nonresidents may be explained by the 
fact that in both Fremont and Richmond these properties are more 
| 
/ 
| 
| 
| 
