Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
Purpose of sampling
Types of sampling
Sample size
Recall… Tradition of Excellence
Statistics
Data Information
5.3
Methods of Collecting Data… Tradition of Excellence
1. Direct Observation
2. Experiments, and
3. Surveys.
5.4
Surveys Tradition of Excellence
•The Response Rate (i.e. the proportion of all people selected who complete
the survey) is a key survey parameter.
5.5
Tradition of Excellence
Population
vs
Sample
Sampling Tradition of Excellence
•In any case, the sampled population and the target population should
be similar to one another.
5.7
Black/Blue/Green/Red
Thin/Bold
• Haphazard sampling
• Quota sampling
• Purposive sampling
• Snowball sampling
• Deviant case sampling
• Sequential sampling
• Theoretical sampling
Haphazard / Accidental / Convenience sample Tradition of Excellence
L JK K J K
K L L J K
L K K K K K L LL J K L J J
J K K K K L L L L J L
L L KK L K
K K K
K K J
K L L K K K K KK LL K
L J K LK
L J K K
K LL K K K L LL K
J JJ L K
K L KKKL KL K KL K K J K
J L K L
K K K KK K KL K L L KL L L K L K
K KK K K L L K K K L K J J J K
K L L
K K KK K KL K K K
KL K L L K J K KL L KL J K
K K K K K K K K K
L JK L KKK K L KKK K L L L
K
L K K K K K L L JK
K K
K K K L J K L L
L K K K K L K L K
L
L K
J L J L
L K K K K KK K K K K K L L L K K
J K K L K K K KK K K K L K K J L K L
K L L K K
L K K
K K K KK K K K L K LK K
K K K J KKK K K JK K K K K L
L K L K K L K
K K K K L K KL J
L K K K K K K KK K K
L L K L K K K K J L K L
KK
K
K
K K K J KK K K
K K K K
K K LK K K L K K K KK K
K K L K K K
K
J
Quota sample Tradition of Excellence
L JK K J J J J J
K L L J
L K K K K K L LL J K L J J K K K K
J K K K K L L L L J L L L L
L L KK L K
K K
K K J J J J J
K L L K K K K KK LL K
L J K LK
L J K
K LL K K K L LL K
J JJ L K K K K K
K L KKKL KL K KL K K J L K L
J L L L L
K K K KK K KL K L L KL L L K L J J
K KK K K L L K K K L K J J J J J
K L L K K K K
K K KK K KL K K K
KL K L L K J K KL L KL J
K K K K K K K K L L L L
L JK L KKK K L KKK K L L L J J J J
L K K K K K L L JK
K K
K K K L J K L K K K K
L K K K K L K L K
L
L K
J L J L
L K K K K KK K K K K K L L L K L L L L
J K K L K K K KK K K K L K K J L K
K L L K K
L K K
K K K KK K K K L K LK
K K K J KKK K K JK K K K K
L K L K K L K
K K K K L K KL
L K K K K K K KK K K
L L K L K K K K J L KK K
K
K
K K K J KK K
K K K K
K K LK K K L K K K KK
K K L K K
K
Purposive sample Tradition of Excellence
L JK K J
K L L J
L K K K K K L LL J K L J J
J K K K K L L L L J
L L KK L K
K K
K K J
K L L K K K K KK LL K
L J K LK
L J K
K LL K K K L LL K
J JJ L K J J
K L KKKL KL K KL K K J L K L
J
L KL L L K L J J
K K K KK K KL K L J J J
K KK K K L L K K K K L L L K J J
K K KK K KL K K K
KL K L L K J K KL L KL J J J
K K K K K K K K
L JK L KKK K L KKK K L L L
L K K K K K L L JK
K K
K K K L J K L
L K K K K L K L K
L
L K
J L J L
L K K K K KK K K K K K L L L K
J K K L K K K KK K K K L K K J L K
K L L K K
L K K
K K K KK K K K L K LK
K K K J KKK K K JK K K K K
L K L K K L K
K K K K L K KL
L K K K K K K KK K K
L L K L K K K K J L KK K
K
K
K K K J KK K
K K K K
K K LK K K L K K K KK
K K L K K
K
Snowball sample Tradition of Excellence
L JK K J
K L L J
L K K K K K L LL J K L J J
J K K K K L L L L J
L L KK L K
K K
K K J
K L L K K K K KK LL K
L J K LK
L J K
K LL K K K L LL K
J JJ L K
K L KKKL KL K KL K K J L K L
J
K K K KK K KL K L L KL L L K L
K KK K K L L K K K L K J J J
K L L
K K KK K KL K K K
KL K L L K J K KL L KL J
K K K K K K K K
L JK L KKK K L KKK K L L L
L K K K K K L K J K L JK
K K L K L
L K K K K L K L K
L
L K
K J L J L
L K K K K KK K K K K K L L L K
J K K L K K K KK K K K L K K J L K
K L L K K
L K K
K K K KK K K K L K LK
K K K J KKK K K JK KK K K K
L K L K K L K K K L K KL
K K K K K K KK K K
L L L K L K K K K K J L K
KK
K
K
K K K J KK K
K K K K
K K LK K K L K K K KK
K K L K K
K
Deviant case sampling Tradition of Excellence
L JK K J
K L L J
K K K K J K L J
L K
J
L LL
L L J
J K
L KK L K K K K K L L
K L K K K J
K L L K K K K KK LL K
L K LK
K LL K K K L L
J
J JJ L K
J K K
LL K
K L KKKL KL K KL K K J L K
J
K K K KK K KL K L L KL L L K L
L
J J J
K
K KK K K L L K K K K L L L K
K K KK K KL K K K
KL K L L K J K KL L KL J
K K K K K K K K
L JK L KKK K L KKK K L L L
L K K K K K L L JK
K K
K K K L J K L
L K K K K L K L K
L
L K
J L J L
L K K K K KK K K K K K L L L K
J K K L K K K KK K K K L K K J L K
K L L K K
L K K
K K K KK K K K L K LK
K K K J KKK K K JK K K K K
L K L K K L K
K K K K L K KL
L K K K K K K KK K K
L L K L K K K K J L KK K
K
K
K K K J KK K
K K K K
K K LK K K L K K K KK
K K L K K
K
Theoretical sampling Tradition of Excellence
Other demos : 1, 2
Probability sampling techniques Tradition of Excellence
A link
Simple random sample
Tradition of Excellence
1 2 3 4 5 6 7 8 9 10
11 12 13 14 15 16 17 18 19 20
21 22 23 24 25 26 27 28 29 30
31 32 33 34 35 36 37 38 39 40
41 42 43 44 45 46 47 48 49 50
51 52 53 54 55 56 57 58 59 60
61 62 63 64 65 66 67 68 69 70
71 72 73 74 75 76 77 78 79 80
81 82 83 84 85 86 87 88 89 90
91 92 93 94 95 96 97 98 99 100
101 102 103 104 105 106 107 108 109 110
111 112 113 114 115 116 117 118 119 120
121 122 123 124 125 126 127 128 129 130
131 132 133 134 135 136 137 138 139 140
141 142 143 144 145 146 147 148 149 150
151 152 153 154 155 156 157 158 159 160
161 162 163 164 165 166 167 168 169 170
171 172 173 174 175 176 177 178 179 180
181 182 183 184 185 186 187 188 189 190
191 192 193 194 195 196 197 198 199 200
Systematic sample
Tradition of Excellence
1 2 3 4 5 6 7 8 9 10
11 12 13 14 15 16 17 18 19 20
21 22 23 24 25 26 27 28 29 30
31 32 33 34 35 36 37 38 39 40
41 42 43 44 45 46 47 48 49 50
51 52 53 54 55 56 57 58 59 60
61 62 63 64 65 66 67 68 69 70
71 72 73 74 75 76 77 78 79 80
81 82 83 84 85 86 87 88 89 90
91 92 93 94 95 96 97 98 99 100
101 102 103 104 105 106 107 108 109 110
111 112 113 114 115 116 117 118 119 120
121 122 123 124 125 126 127 128 129 130
131 132 133 134 135 136 137 138 139 140
141 142 143 144 145 146 147 148 149 150
151 152 153 154 155 156 157 158 159 160
161 162 163 164 165 166 167 168 169 170
171 172 173 174 175 176 177 178 179 180
181 182 183 184 185 186 187 188 189 190
191 192 193 194 195 196 197 198 199 200
Stratified sampling Tradition of Excellence
Cluster sampling Tradition of Excellence
How large should a sample be?
Tradition of Excellence
5.24
Sampling and Non-Sampling Errors… Tradition of Excellence
•Two major types of error can arise when a sample of observations is taken from a
population:
•sampling error and nonsampling error.
•Sampling error refers to differences between the sample and the population that
exist only because of the observations that happened to be selected for the
sample. Random and we have no control over.
•Nonsampling errors are more serious and are due to mistakes made in the
acquisition of data or due to the sample observations being selected improperly.
Most likely caused be poor planning, sloppy work, act of the Goddess of Statistics,
etc.
5.25
Sampling Error… Tradition of Excellence
5.26
Nonsampling Error… Tradition of Excellence
•Nonsampling errors are more serious and are due to mistakes made in
the acquisition of data or due to the sample observations being
selected improperly. Three types of nonsampling errors:
•Note: increasing the sample size will not reduce this type of error.
5.27
Errors in data acquisition… Tradition of Excellence
5.28
Nonresponse Error… Tradition of Excellence
•…refers to error (or bias) introduced when responses are not obtained
from some members of the sample, i.e. the sample observations that
are collected may not be representative of the target population.
•As mentioned earlier, the Response Rate (i.e. the proportion of all
people selected who complete the survey) is a key survey parameter
and helps in the understanding in the validity of the survey and sources
of nonresponse error.
5.29
Selection Bias… Tradition of Excellence
•…occurs when the sampling plan is such that some members of the
target population cannot possibly be selected for inclusion in the
sample.
5.30
Group work Tradition of Excellence
6.5 10
P( Z > )=
3. 3
P( Z > 1.06 ) = 85%
study 1: 50 cases, 50 controls
Critical value=
0+10*1.96=20
Tradition of Excellence
Za/2=1.96
2.5% area
Power closer to
15% now.
Study 2: 18 treated, 72 controls, STD DEV = 2
Clinically relevant
alternative: Power is nearly
difference=4 points 100%!
Study 2: 18 treated, 72 controls, STD DEV=10
Tradition of Excellence
Critical value=
0+0.52*1.96 = 1
Clinically relevant
alternative:
difference=1 point
Sample size calculations Tradition of Excellence
Represents the
desired power
Sample size in each (typically .84 for 80%
group (assumes equal power).
sized groups)
2s ( Z b + Za/2 )
2 2
n= 2
difference Represents the desired
Standard deviation of level of statistical
the outcome variable Effect Size (the
difference in significance (typically
means) 1.96).
Simple formula for difference in proportions
Tradition of Excellence
Represents the
desired power
Sample size in each (typically .84 for 80%
group (assumes equal power).
sized groups)
n=
(p1 - p2 ) 2
Represents the
A measure of Effect Size (the desired level of
variability (similar to difference in statistical
standard deviation) proportions) significance
(typically 1.96).
SAMPLE SIZE AND POWER FORMULAS
-0
e.g. here :Z b = ; power = 50%
standard error (diff)
Sample size formula for difference in means
Tradition of Excellence
(r + 1) s ( Z power + Za/2 )
2 2
n1 = 2
r difference
where :
n 1 = size of smaller group
r = ratio of larger group to smaller group
s = standard deviation of the characteristic
diffference = clinically meaningful difference in means of the outcome
Z power = corresponds to power (.84 = 80% power)
Za / 2 = corresponds to two - tailed significance level (1.96 for a = .05)
Derivation of a sample size formula: Tradition of Excellence
p(1 - p) p(1 - p)
+
n1 n2
Derivation of a sample size formula: Tradition of Excellence
2
r +1 p (1 - p )( Z power + Z a /2 )
n=
r ( p1 - p 2 ) 2
General sample size needs when outcome is
binary: Tradition of Excellence
2
r + 1 p(1 - p)( Z b + Za / 2 )
n=
r ( p1 - p2 ) 2
where :
n = size of smaller group
r = ratio of larger group to smaller group
p1 - p2 = clinically meaningful difference in proportions of the outcome
Z b = corresponds to power (.84 = 80% power)
Za / 2 = corresponds to two - tailed significance level (1.96 for a = .05)
Compare with when outcome is
continuous: Tradition of Excellence
(r + 1) s ( Z b + Za/2 )
2 2
n1 = 2
r difference
where :
n1 = size of smaller group
r = ratio of larger group to smaller group
s = standard deviation of the characteristic
diffference = clinically meaningful difference in means of the outcome
Z b = corresponds to power (.84 = 80% power)
Za / 2 = corresponds to two - tailed significance level (1.96 for a = .05)
Sample size for paired data:
Tradition of Excellence
2
s d ( Z b + Za/2 ) 2
n= 2
difference
where :
n = sample size
s = standard deviation of the within - pair difference
diffference = clinically meaningful difference
Z b = corresponds to power (.84 = 80% power)
Za / 2 = corresponds to two - tailed significance level (1.96 for a = .05)
Paired data difference in proportion: sample
size: Tradition of Excellence
2
p(1 - p)( Z b + Za / 2 )
n=
( p1 - p2 ) 2
where :
n = sample size for 1 group
p1 - p2 = clinically meaningful difference in dependent proportions
Z b = corresponds to power (.84 = 80% power)
Za / 2 = corresponds to two - tailed significance level (1.96 for a = .05)