Probability Sampling

Types of Probability
Sampling Designs

 Simple random sampling

 Stratified sampling
 Systematic sampling
 Cluster (area) sampling
 Multistage sampling
 Some Definitions
 N = the number of cases in the sampling
frame
 n = the number of cases in the sample
NCn = the number of combinations (subsets)


of n from N
 f = n/N = the sampling fraction
Simple Random Sampling
• Objective: Select n units out of N such that
every NCn has an equal chance.
• Procedure: Use table of random numbers,
computer random number generator or
mechanical device.
• Can sample with or without replacement.
• f=n/N is the sampling fraction.
Simple Random Sampling

Example:
 Small service agency.
 Client assessment of quality of service.
 Get list of clients over past year.
 Draw a simple random sample of n/N.
Simple Random Sampling

List of clients
Simple Random Sampling

List of clients

Random subsample
Stratified Random Sampling

• Sometimes called "proportional" or

"quota" random sampling.
• Objective: Population of N units divided
into nonoverlapping strata N1, N2, N3, ...
Ni such that N1 + N2 + ... + Ni = N; then
do simple random sample of n/N in
each strata.
Stratified Sampling - Purposes:
• To insure representation of each strata,
oversample smaller population groups.
• Administrative convenience -- field
offices.
• Sampling problems may differ in each
stratum.
• Increase precision (lower variance) if
strata are homogeneous within (like
blocking).
Stratified Random Sampling

List of clients
Stratified Random Sampling

List of clients

African-American
Hispanic-American Other
s

Strata
Stratified Random Sampling
List of clients

African-American Hispanic-American Others

Strata

Random subsamples of n/N

 Proportionate vs.
Disproportionate Stratified
Random Sampling

 Proportionate: If sampling fraction is equal for

each stratum
 Disproportionate: Unequal sampling fraction
in each stratum
 Needed to enable better representation of
smaller (minority groups)
Systematic Random Sampling

Procedure:
 Number units in population from 1 to N.
 Decide on the n that you want or need.
 N/n=k the interval size.
 Randomly select a number from 1 to k.
 Take every kth unit.
 Systematic Random
Sampling

 Assumes that the population is randomly

ordered.
 Advantages: Easy; may be more precise than
simple random sample.
Systematic 1
2
26
27
51
52
76
77

Random
3 28 53 78
4 29 54 79
5 30 55 80

Sampling 6
7
8
31
32
33
56
57
58
81
82
83
N = 100 9
10
34
35
59
60
84
85
11 36 61 86
12 37 62 87
13 38 63 88
14 39 64 89
15 40 65 90
16 41 66 91
17 42 67 92
18 43 68 93
19 44 69 94
20 45 70 95
21 46 71 96
22 47 72 97
23 48 73 98
24 49 74 99
25 50 75 100
Systematic 1
2
26
27
51
52
76
77

Random
3 28 53 78
4 29 54 79
5 30 55 80
N = 100
Sampling
6 31 56 81
7 32 57 82
8 33 58 83
9 34 59 84
10 35 60 85
Want n = 20 11 36 61 86
12 37 62 87
13 38 63 88
14 39 64 89
15 40 65 90
16 41 66 91
17 42 67 92
18 43 68 93
19 44 69 94
20 45 70 95
21 46 71 96
22 47 72 97
23 48 73 98
24 49 74 99
25 50 75 100
Systema 1
2
26
27
51
52
76
77
tic 3
4
28
29
53
54
78
79

Random N = 100
5
6
30
31
55
56
80
81
7 32 57 82
Samplin 8
9
33
34
58
59
83
84
g want n = 20 10
11
35
36
60
61
85
86
12 37 62 87
13 38 63 88
N/n = 5 14 39 64 89
15 40 65 90
16 41 66 91
17 42 67 92
18 43 68 93
19 44 69 94
20 45 70 95
21 46 71 96
22 47 72 97
23 48 73 98
24 49 74 99
25 50 75 100
 1 26 51 76
2 27 52 77
3 28 53 78
Systematic 4
5
29
30
54
55
79
80

Random N = 100 6
7
8
31
32
33
56
57
58
81
82
83
Sampling Want n = 20
9
10
34
35
59
60
84
85
11 36 61 86
12 37 62 87
13 38 63 88
N/n = 5 14 39 64 89
15 40 65 90
16 41 66 91
Select a random number from 1-5: 17 42 67 92
18 43 68 93
chose 4 19 44 69 94
20 45 70 95
21 46 71 96
22 47 72 97
23 48 73 98
24 49 74 99
25 50 75 100
Systemat 1
2
26
27
51
52
76
77
3 28 53 78
ic 4
5
29
30
54
55
79
80

Random N = 100
6
7
31
32
56
57
81
82
8 33 58 83
Sampling Want n = 20
9
10
34
35
59
60
84
85
11 36 61 86
12 37 62 87
13 38 63 88
N/n = 5 14 39 64 89
15 40 65 90
16 41 66 91
Select a random number from 1-5: 17 42 67 92
18 43 68 93
chose 4 19 44 69 94
20 45 70 95
21 46 71 96
22 47 72 97
Start with #4 and take every 5th unit 23 48 73 98
24 49 74 99
25 50 75 100
Cluster (Area) Random
Sampling
Procedure:
 Divide population into clusters.
 Randomly sample clusters.
 Measure all units within sampled
clusters.
 Cluster (Area) Random
Sampling
 Advantages: Administratively useful,
especially when you have a wide
geographic area to cover.

 Examples: Randomly sample from city

blocks and measure all homes in
selected blocks.
Multi-Stage Sampling
 Cluster (area) random sampling can
be multi-stage.
 Any combinations of single-stage
methods.
Multi-Stage Sampling
Example: Choosing students from schools
 Select all schools; then sample within
schools.
 Sample schools; then measure all students.
 Sample schools; then sample students.