Page 376 - Invited Paper Session (IPS) - Volume 2

IPS280 VEUN, Thy et al.
sizes calculated from the three methods above were multiplied by the
design effect of 2 in this study. The deff of 2 was used because the study
on deff in each province on some indicators from the CSES2014, such as,
the number of household in each village, total household income and
consumption, showed that the number of households in each village
exhibited deff ranging from 0.3 to 1.8 while total expenditure and total
income variables have their deff ranging from 0.5 to 1.7 and from 0.5 to
1.8 respectively.

3. Main finding/results
The required sample sizes of households in each province with a further
stratification within each province and the without-further stratification are
shown in the Table 1. Results show that within 5 percent level of precision, very
huge reduction in the required sample sizes of households for each province
was achieved in the case when a further stratification within each province was
done compared to the one without a further stratification. Without a further
stratification, the total sample size of 42,515 households is required for whole
country, whereas only 8,340, 10,290, and 14,670 households are required,
respectively, for LHK, CDH, and GGH under the case when a further
stratification within each province was done. Among these three methods, LHK
method provided the smallest sample size of households followed by the CDH
method. The GGH method of stratum boundary determination gave the
largest sample size. To gain more insight into how these three methods have
worked for this studied population, all the three levels of required precision
used are showed in the Table 2. It was found that LHK was still the method
that provided the smallest sample size among the three, followed by the CDH
method, in all levels of required precision. The GGH method performed the
least because it provided the largest sample size requirement in all required
levels of precision.

Table 1. The required sample sizes of households in each province under
5% CV without and with further stratification within province
Total Without further With further stratification
Province Name number of stratification within within province
households province
LHK CDH GGH
Banteay 153,396 2,032 420 450 750
Meanchey
Battambang 226,460 1,233 330 330 630
Kampong Cham 222,380 1,094 300 300 570
Kampong 107,875 1,099 300 390 540
Chhnang
Kampong Speu 157,207 851 270 330 480

363 | I S I W S C 2 0 1 9