Scribd
Upload
523 views

Multistage Sampling

Multistage sampling involves selecting samples in multiple stages. In a two-stage sample, primary sampling units (PSUs) are selected in the first stage, and secondary sampling units (SSUs) are selected from within each selected PSU in the second stage. The variance of an estimator from a two-stage sample depends on variability between PSUs and between SSUs within PSUs. Optimal sample sizes aim to minimize variance for a given cost by balancing the number of PSUs and SSUs per PSU selected.

Uploaded by

amryina
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
523 views

Multistage Sampling

Multistage sampling involves selecting samples in multiple stages. In a two-stage sample, primary sampling units (PSUs) are selected in the first stage, and secondary sampling units (SSUs) are selected from within each selected PSU in the second stage. The variance of an estimator from a two-stage sample depends on variability between PSUs and between SSUs within PSUs. Optimal sample sizes aim to minimize variance for a given cost by balancing the number of PSUs and SSUs per PSU selected.

Uploaded by

amryina
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 29

Multistage Sampling

Outline
Ciri-ciri Contoh Reka Bentuk Multi- peringkat
Kebarangkalian pemilihan dalam persampelan
pelbagai peringkat
Anggaran parameter
Pengiraan ralat piawai
Kecekapan sampel pelbagai peringkat
2

Introduction
Persampelan pelbagai peringkat bermakna apa
namanya -> terdapat pelbagai peringkat dalam
proses pensampelan
Bilangan peringkat boleh menjadi banyak, walaupun
ia jarang sekali mempunyai lebih daripada 3
Untuk topik ini kita akan menumpukan perhatian
kepada pensampelan dua peringkat
Juga dikenali sebagai subsampling
3

Persampelan Unit dalam Multi- peringkat


Persampelan
First-stage sampling units are called primary
sampling units or PSUs.
Second-stage sampling units are called
secondary sampling units or SSUs.
Last-stage sampling units are called ultimate
sampling units or USUs.
4

4-stage Sampling (example)


Villages

EAs

Dwelling

Persons
A

C
5

Your Examples
Estimation Domains
Stratification
Number of stages
Sampling units for each stage
Sample selection scheme in each stage
Sampling frames used in each stage
6

Example: Maldives HIES 2002

Two-Stage Sampling
Stage One. Select sample of clusters from population
of clusters.
Using any sampling scheme, usually: SRSWOR, PPSWR,
LSS

Stage Two. Select sample of elements within each of


the sample clusters.
Language: also referred to as subsample of elements within
a cluster
Subsampling can be done also using any sampling scheme
8

Most Large-Scale Surveys Use


Multi-stage Sampling Because
Sampling frames are available at higher stages but not for
the ultmate sampling units. Construction of sampling
frames at each lower stage becomes less costly.
Cost efficiency with use of clusters at higher stages of
selection
Flexibility in choice of sampling units and methods of
selection at different stages
Contributions of different stages towards sampling variance
may be estimated separately

Probabilities of Selection
Probability that an element in the population is
selected in a 2-stage sample is the product of
Probability that the cluster to which it belongs is
selected at the first stage
Probability that the element is selected at the second
stage given that the cluster to which it belongs is
selected at the first stage
10

Example: Two-Stage Samples

11

Estimation Procedures: Illustrations

SRS at stage 1 and SRS at stage 2


SRS at stage 1 and LSS at stage 2 (b from B)
PPSWR at stage 1 and SRS at stage 2 (b from B)

12

SRS SRS: Estimation of Total


Estimator of Total

Y2 s

A a A a
Y B y
a 1
a 1

Variance of Estimator

1
Y Y
Sb

A 1 1
2

A
1
1
A
1 2

2
2
2 1

V Y A Sb B S
a 1 b B
a A

1
Y Y
S

B 1 1
2

13

SRS SRS: Variance of Estimator

A
1
1
A
1 2

2
2
2 1

V Y A Sb B S
a 1 b B
a A

Sources of Variation = {PSUs} + {SSUs}


Total variability = Variability among PSUs + Variability of SSUs

14

SRS-SRS: Estimating Variance


Estimator of Variance of Estimator for Total

a
1
1
A
1 2

2
2
2 1

v Y A sb B s
a 1 b B
a A

1
1

sb
Y Y

a 1 1
a 1
2

1
1
s
y

b 1 1
b
2

y
1

15

SRS-SRS: Estimating a Mean


Each PSU has same number of elements, B
Subsample of b elements is selected
a

Y y
a 1
where

2
b

2
w

a S

b S

V Y 1 1
A a

B ab
A

1
S
Y Y

A 1 1
A
B
2
1
1
2
2
2
S w S ; S
Y Y

A 1
B 1 1
2
b

16

with variance estimate

v Y

2
s
a sb2
b

1 1 w
A a
B ab

s
y

a 1 1
2
b

a
b
2
1
1
2
2
2
sw s ; s
y y

a 1
b 1 1

17

SRS-SRS: Population Mean (1)


PSUs have unequal sizes
a
1
Y1
B y

aB 1
2
A
B
1
1
1
1
1

2
2

V (Y1 ) ( ) S1b
(

)
S

2
a A
aA 1 B b B

B 1 1 2
1 1 2
1

v(Y1 ) ( ) s1b
( ) s

a A
aA 1 B b B
18

SRS-SRS: Population Mean (2)


PSUs have unequal sizes
a
1
Y2 y
a 1
A
1

bias(Y2 )
( B B )Y

AB 1

1 1 2
1
1
1 2

V (Y2 ) ( ) S 2b
( ) S

a A
aA 1 b B
19

SRS-SRS: Population Mean (3)


PSUs have unequal sizes
a

Y3

B
1
a

B
1

A
1
1
1
1 2

2
2 1
V (Y3 ) ( ) S3b 2 B ( ) S
a A
B aA 1
b B

20

SRS-LSS: Estimation of Mean


a
1
Y y
a 1

A
1
1
1
1
1

V Y sb2 1 S2 1 b 1
ab
a A
B A 1

2
b

v Y
a

21

PPSWR-SRS: Estimation of Total

Y
1
Y
a 1 p

; Y B y

p
1

v Y
a a 1

22

Design Effect for 2-stage Sample


)
V
(
Y
clu
DEFFclu (Y )
1 ( B 1)
Vsrs (Y )

If is positive, the design effect decreases as


the subsample size b decreases.
For fixed n=ab, the smaller the sub-sample size
and, hence, the larger the number of clusters
included in the sample, the more precise is the
sample mean.
23

Designing a Cluster Sample


What overall precision is needed?
What size should the psus be?
How many ssus should be sampled in each psu
selected for the sample?
How many psus should be sampled?

24

Choosing psu Size


Often a natural unit not much choice
Larger the psu size, more variability within a psu
ICC is smaller for large psu compared to small psu
but, if psu size is too large, less cost efficient

Need to study relationship between psu sizes


and ICC and costs
25

Optimum Sample Sizes (1)


Goal: get the most information (and hence, more
statistically efficient) for the least cost
Illustrative example: PSUs with equal sizes,
SRSWOR at both stages

26

Optimum Sample Sizes (2)


Variance function
Cost function

2
b

2
w

S
S
V

a ab
C C0 ac1 abc2

Minimize V subject to given cost C*

27

Optimum Sample Sizes (3)


Minimize V subject to given cost C*
Optimum a=a* and b=b*

b* c1S w2 / c2 Sb2

a*

2
S
*
(C C0 ) b
c1

c S c2 S
2
1 b

2
w

28

Optimum Sample Sizes (4)


Optimum b=b*

c1 S
c1 1 roh
b*

c2 S
c2 roh
2
w
2
b

29

You might also like