Page 44 - Agricultural Census 2000 NWFP Vol II Part 1

culminating at the varied not, however, be presumed that
estim·ates. In addition to the all the national .· and provincial
above, another _ major factor totals · are subjected to a
seems to be the non-coverage substantial rounding effect. As
of migratory herds by the in the large aggregates, vide the
Patwaris in Agricultural law of inertia of large numbers,
Censuses whereas the Stock such effects cancel · out each
Assistants take due care for other thus curtailing rounding
their enumeration in Livestock effect . at the national or
Censuses. It may be noted that provincial levels to a minimum.
the data from Agricultural
Census are primarily meant for
studying relationships between
livestock population of different
categories · and sizes of farms.
Therefore, for studying
livestock population, their,
products - and for developing
their growth rates, data
provided by _ Livestock Census
may be preferred over the
Agricultural Census data.

7. In some cases, the figures in a
row or a column may not add up
exactly to their respective row or
column total. These differences
are due to rounding which was
unavoidable in view of the_ fact
that each area figure was
collected in terms- of acres and

kanals and converted into acres
· and hectares. Further, each
data item was multiplied with its
raising factor worked out upto
three places of decimal. The
size of rounding effect in a total
normally depends upon the
number of independently
worked out components
included therein. Consequently,
the rounding effect will be larger
in a national or provincial total
than in a district total. It should

( XLII)