12.02.2013 Views

PALESTINIAN SOCIETY - Fafo

PALESTINIAN SOCIETY - Fafo

PALESTINIAN SOCIETY - Fafo

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

Having thus decided the number of sample cells in each of the<br />

PSUs it remains to determine the allocation of the PSU household<br />

sample among cells. According to (3.6) this allocation obviously has<br />

to beproportionate inordertohave alocal(withinPSU) epsemdesign.<br />

As stated previously, the latter requires R(s,k,c) to be a constant<br />

independent of the cell (c) under consideration. TIms, in R( s,k,e) we<br />

may ornit the index c and reformulate (3.6):<br />

B (5, k)<br />

(3.7) d (5, k, c) = R (5, k) b (5, k) D (5, k, c)<br />

In order to calculate R(s,k) foreach of the sample PSUs we take the<br />

sum for every (s,k) of both sides of (3.7):<br />

b(s.k)<br />

B (5, k)<br />

b(s. k)<br />

(3.8) c� d (5, k, c) = R (5, k) b (5, k) c� D (5, k, c)<br />

The left hand side adds to the PSU sample size, d(s)lk(s). On the<br />

righ t hand side all statisties are known except for the constant R( s,k).<br />

Hence R(s,k) is determined, and the individual d(s,k,c)s can be<br />

calculated from (3.7), concluding the sample allocation calculations.<br />

The d(s,k,c)s arrived at also determine the number of housing units<br />

selected from each celI.<br />

The fmal household and cell sample allocation is displayed in table<br />

A.9 for the selected PSUs.<br />

Before concluding this section, we retum to the overall inclusion<br />

probability (3.3), to see how this can be calculated.<br />

. 5" c, , -<br />

(3 3) P ( k h ) _ k (5) D (5, k) b (5, k) d (5, k, c) 1<br />

D (5) 8 (5, k) H (5, k, c,) D (5, k, c, h)<br />

The first fraction on the right hand side is the 1st stage inclusion<br />

probability. At the planning stage the numbers of PSU and stratum<br />

households, the D(s,k)s and D(s)s, were not available. Instead, the<br />

Benvenisti estimates6 of the total population figures were used<br />

(equation (3.1» . The second fraction is the 2nd stage inclusion<br />

probability which can be calculated from the figures in table A.9. To<br />

calculate the third fraetion, the d(s,k,c)s are taken from the fmally<br />

observed (net) sample, and the H(s,k,c)s are estimated by formula<br />

(2.4) in the Gaza design section. The last fraction - the 4th stage<br />

inclusion probability -isdetermined by the sample observations ofthe<br />

D(s,k,c,h)s.<br />

329

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!