Professional Documents
Culture Documents
For
First Year Medical Students
1. Define sampling
2. Enumerate the advantages of sampling
3. Enumerate and discuss the uses of sampling
4. Define some terms used in sampling
a. Population and sample
b. Target population and sampling population
c. Sampling unit and elementary unit
d. Sampling frame
5. Enumerate and discuss the two types of
sampling designs
6. Enumerate the criteria of a good sampling
design
Most of the time is not
feasible
Difficult
Expensive
- is the act of studying or examining
only a segment of the population to
represent the whole
.25 570 476 374 416 336 261 346 274 198
.30 396 331 260 289 234 174 241 190 137
.40 223 186 146 162 131 98 135 107 77
.80 56 47 36 41 33 25 34 27 19
.90 44 37 29 32 26 19 27 21 15
1.00 36 30 23 26 21 16 22 17 12
•E/S is the standardized effect size, computed as E (expected effect size) divided by S (standard deviation of
the outcome variable)
Smaller of Upper number: = 0.05 (one-tailed) or = 0.10 (two-tailed); = 0.20
P1 and P2* Middle number: = 0.025 (one-tailed) or = 0.05 (two-tailed); = 0.20
Lower number: = 0.025 (one-tailed) or = 0.05 (two-tailed); = 0.10
Expected difference between P1 and P2
0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50
0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50
0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50
0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50
.90 342 - - - - - - - - -
434 - - - - - - - - -
581 - - - - - - - - -
•P1 represents the proportion of subjects expected to have the outcome in one group; P2 in the other group.
(in a case-control study, P1 represents the proportion of cases with the predicted variable; P2 the proportion
of controls with the predictor variable)
One-tailed = 0.005 0.025 0.05
Two-tailed = 0.10 0.05 0.010
r2 = 0.05 0.10 0.20 0.05 0.10 0.20 0.05 0.10 0.20
0.05 7118 5947 4663 5193 4200 3134 4325 3424 2469
0.10 1773 1481 1162 1294 1047 782 1078 854 616
0.15 783 655 514 572 463 346 477 378 273
0.20 436 365 287 319 259 194 266 211 153
0.25 276 231 182 202 164 123 169 134 98
0.30 189 158 125 139 113 85 116 92 67
0.50 62 52 42 46 38 29 39 31 23
0.60 40 34 27 30 25 19 26 21 16
0.70 27 23 19 20 17 13 17 14 11
0.80 18 15 13 14 12 9 12 10 8
Confidence Level
W/S * 90% 95% 99%
0.10 1083 1537 2665
0.15 482 683 1180
0.20 271 385 664
0.40 68 97 166
0.50 44 62 107
0.60 31 43 74
.70 23 32 55
0.80 17 25 42
0.90 14 19 33
1.00 11 16 27
• W/S is the standardized width of the confidence interval, computed as W (desired total width) divided by S (standard
deviation of the variable)
Upper number: 90% confidence level
Middle number: 95% confidence level
Lower number: 99% confidence level
Expected Total width of confidence interval (W)
proportion (P) 0.10 0.15 0.20 0.25 0.30
0.10 98 - - - -
139 - - - -
239 - - - -
0.15 138 62 - - -
196 88 - - -
339 151 - - -
0.20 174 77 43 - -
246 110 62 - -
425 189 107 - -
0.25 203 91 51 33 -
289 128 73 47 -
498 221 125 80 -
Upper number: 90% confidence level
Middle number: 95% confidence level
Lower number: 99% confidence level
Expected Total width of confidence interval (W)
proportion (P) 0.10 0.15 0.20 0.25 0.30
0.30 228 101 57 37 26
323 144 81 52 36
558 248 139 90 62
OR X P2 ÷ (1 – P2 + OR X P2)
Problem:
The researcher question is whether elderly smokers have a
greater incidence of skin cancer than nonsmokers. A review
of previous literature suggests that the 5 year incidence of
skin cancer is about 0.20 in elderly nonsmokers. At an α
(one-tailed) = 0.05 and power = 80%, how many smokers
and nonsmokers will need to be studied to determine
whether the 5 year skin cancer incidence is at least 0.30 in
smokers ? Why was a one-tailed alternative hypothesis
chosen ?
1. Ho.: The incidence of skin cancer is the same in elderly
smokers and nonsmokers
Answer: 664
To estimate the sample size for that confidence interval,
the investigator must:
1. Estimate the expected proportion with the variable of interest in
the population ( If more than half of the population is expected
to have the characteristic, then plan sample size based on the
proportion expected not to have the characteristic)
Answer: 139
1. Use continuous variables
- This permits smaller sample sizes than dichotomous variables
5. If the level of significance of the test is changed from .05 to .01, will the sample
size needed in the study increase or decrease?
1. Quota sampling and purposive sampling are examples of
probability procedures.