Page 25 - Agricultural Census 2000 NWFP Vol II Part 1

sampled, a random number over and . were divided into convenient
zero and upto 0.999 was selected . number of sectors, each having
. The first patwar circle with cumulative approximately 500-700 households.
probability equal to or greater than the The number of households in a sector
random number . was considered was treated as the measure of size to
selected. Similarly, random number work out the probability of selection of
· plus 1 (one) · selected the second a sector. One sector from each big
patwar circle, random number plus 2 mouza was · selected with equal
(two) selected the third patwar circle probability.
and so on until . the pre-determined
number · of patwar circles were THIRD STAGE; Selection of
selected from the sub-division. Clusters of Households:

SECOND STAGE; Selection of A . systematic selection of
Mouzas: clusters of ne rly 30 households each
9
was made wit~ equal probability
It was, a priori, decided to selecJ from a!I the selected mouzas with the
two mouzas per selected patwar' help of the list of households (Form-
circle. The mouzas falling in certainty 1 ). The number of clusters to be
and selected · patwar circles were selected, with a minimum of two, was
arranged in serpentine fashion, i.e. determined \Nith regard to raising
within the odd numbered kanungoes factor of the mouza and the pre-fixed
in descending order of cultivated area overali raising factor at cluster level
of the mouzas and within even for a particular mouza. The limit of
numbered kanungoes in ascending overall raising factor was fixed at 25
order of cultivated area of mouzas. for all the rural settled . · areas.
The method employed at the first Ultim~teiy the sample included:
stage was repeated at patwar circle
level for segregating the certainty i) The households in the selected
• mouzas and also for the selection of clusters.
the non-certainty mouzas. Like the
patwar circles with small probabiiity of . ii) The househoids located in the
selection, the mouzas getting small selected urban blocks.
probabilities · were merged together
by applying the principles mentioned iii) The households located in the
in the first stage and treated as one · selected vilfages I basties / killies
mouza for sampling purposes. of unsettled areas.

One sub-stage at this level was iv) The households located in the
introduc{;;d only for those selected selected mouzas designated as
mouzas which were having nine Mouza Certainty Holdings.
hundred or above households as per v) The households located any
1998 Population · Census. Such where In the country included in

mouzas were termed as big rnouzas.

( XXIII)