การสุ่มตัวอย่างในงานวิจัยสาธารณสุข
- 3. 3.
- 4. INTERPETAT EMPIRICAL CONCEPTUAL PHASE IVE PHASE PHASE H DESIGN RESEARC Research questions
- 7. 1. 2. 3. 4.
- 8. 1. (Representativeness) 2. (Adequate size)
- 10. • •
- 11. 1. (Non- probability sampling) 1.1 Accidental sampling Quota sampling Purposive sampling Convenience sampling
- 13. Random Sampling (Error) 100% (Sampling error)
- 14. (Probability sampling) Probability Sampling – (bias) – (Sampling error)
- 15. SRS (Simple random sampling, SRS) ” (Sampling frame) EXAMPLE
- 16. (Sampling frame) No Name Address 1 …….. 2 …….. 3 …….. 4 …….. 5 …….. 6 …….. 7 …….. 8 …….. 9 …….. 10 …….. A
- 17. A 1) - - )
- 18. (Systematic random sampling) n N (Sampling interval); k = N/n 1 k
- 19. k = 50/10 , 7, 12, 17, 22, 27, 32, 37, 42, 47
- 20. (Stratified sampling) EX “Child development study in Thailand”
- 21. O Curative units A Promotion/Prevention X Rehabitative units units
- 24. EX: • • 4- 5 ?
- 26. (Stratified three-stage sampling) . 1 . 4 4 4
- 27. 85 85 - ( .) 112 48 64 112 48 64 112 48 64 112 48 64 533 277 256
- 28. - - - - - - ภค า รวม ใ เข เท บ ล น ต ศา นก ต ศา อ เข เท บ ล กรุงเท มห ร พ านค 1,530 1,530 - กล (ย กท าง กเว้น ม.) 2,016 864 1,152 เหนือ 2,016 864 1,152 ต อ งเห ะวันอ กเฉีย นือ 2,016 864 1,152 ใต้ 2,016 864 1,152 รวม 9,594 4,986 4,608
- 29. When choosing a sample size, we must consider the following issues: • Objectives: What population parameters we want to estimate/test hypothesis • Sampling/research design is selected • Degree of accuracy required for the study • Spread/variation (variability) of the population • Response rate, practicality: how hard is it to collect data • Time and money available
- 30. 1) Sample size for Simple Random Sampling To estimate mean 2 2 Z N n = 2 2 2 Z ( N 1) E Z2 2 n E2
- 31. Sample size for Simple Random Sampling To estimate proportion 2 Z NP (1 P ) n = 2 2 Z P (1 P ) ( N 1) E n 2P 1 P Z ( ) n E 2
- 32. 1,628 (Pilot survey) z 2 NP (1 P) 5% z 2 P (1 P ) NE 2 % n = )2 (1.645 (1,6280.2)( .8) )( 0 (1.645 (0.2)( .8) (1,6280.052 )2 0 )( ) n = = 156.53 157 5% 90%
- 33. ? z 2 P (1 P ) 95% E 2 n = 2(P 1) 1 (1.96 )( ) P 2 2 P = ½=0 (0.052 ) n= = 384.16 385
- 34. 1.2
- 35. Equal sample L L N2S2 hh n = h1 L N2E2 NhS2 h1 h nh n L
- 36. Proportional Allocation L N NhS2 h h1 n = L N2E2 NhS2 h h1 Nh nh L Nh h1
- 37. 2) Sample size determination for hypothesis testing 2.1 Sample size determination for the test of one proportion
- 38. Example In a particular province the proportion of pregnant women provided with prenatal care in the first trimester of pregnancy is estimated to be 40% by the provincial department of health. Health officials in another province are interested in comparing their success at providing prenatal care with these figures. How many women should be sampled to test the hypothesis that the coverage rate in the second province is % against the alternative that it is not %? The investigators wish to detect a difference of % with the power of the test equal at % and at
- 39. P : coverage rate Ho: P = . Ha: P . ( . or MINITAB can be used to assist in this sample size determination by selecting Stat > Power and sample size > proportion.
- 41. If alternative values of p is equal to .45, a sample size of 1022 would be needed. If alternative values of p is equal to . , a sample size of would be needed. We choose the large sample size, thus a sample size of 1022 is needed for the study.
- 42. 2.2 Sample size determination for the test of two proportions Two-sided test (Z 2pq Z p2q2 p1q1)2 n = 2 (p2 p1)2
- 43. Example 5 It is believed that the proportion of patients who develop complications after undergoing one type of surgery is % while the proportion of patients who develop complications after a second type of surgery is %. How large should the sample size be in each of the two groups of patients if an investigator wishes to detect, with a power of %, whether the second procedure has a complication rate significantly higher than the first at the % level of significance?
- 44. Use MINITAB, click Stat > Power and sample size > proportion. You would complete the dialog box. You want to test one-sided test, click on the options button and choose less than
- 46. Power and Sample Size Test for Two Proportions Testing proportion = proportion (versus <) Calculating power for proportion Alpha = Sample Target Actual Proportion Size Power Power A sample size of would be needed in each group.
- 47. 2.3 Sample size determination for the tes of one mean Two-sided test 2 2 (Z Z ) n 2 ( 0 1 )2
- 48. Example Consider the cholesterol study. Suppose that the null mean is mg% /ml, the alternative mean is mg%/ml, the standard deviation is , and we wish to conduct a significance test for one-sided test at the % level with a power of %. How large should the sample size be?
- 49. MINITAB> click Stat > Power and sample size > sample Z. You want to test one-sided test, click on the options button and choose greater than
- 51. -Sample Z Test Testing mean = null (versus > nul Calculating power for mean = null Alpha = . Sigma = Sample Target Actual Difference Size Power Power Thus, 96 people are needed. To achieve a power of 90% using a 5% significance level
- 52. 2.4 Sample size determination for the test of two means Two-sided test (Z Z )2( 1 2 2)2 2 2 n = ( 2 1)2
- 53. Example Consider the blood pressure study for drug A users and non-drug A users as a pilot study conducted to obtain parameter estimates to plan for a larger study. We wish to test the hypothesis : = versus : . Determine the appropriate sample size for the large study using a two–sided test with a significance level of . and a power of In the pilot study, we obtained = . , S = . = . ,S
- 54. In the pilot study, we obtained = . ,S = . = . ,S n=( - We would require a sample size of 152 people in each group