(M) Sample Size Determination

Question 1

The formula depends on the following (3):

Answer 1

○ parameter being estimated
○ type of research design
○ Sampling design

Question 2

TOF. The rarer the condition, the larger the sample size.

Answer 2

T

Question 3

TOF. Complex data analysis requires a SMALLER sample size than simple analysis.

Answer 3

F

Question 4

TOF. The more heterogeneous the values of the parameter are, the larger the sample size.

Answer 4

T

Question 5

TOF. In general, longitudinal studies require a larger sample size
than case-control and cross-sectional studies.

Answer 5

T

Question 6

TOF. The higher the level of accuracy and precision desired for the resulting estimates, the smaller the sample size necessary.

Answer 6

F (larger)

Question 7

TOF. When only one item or parameters is to be studied,
sample sizes are estimated separately for each important item part or parameter.

Answer 7

F (1 or more item)

Question 8

study ESTIMATION OF THE POPULATION MEAN

Answer 8

OKS

Question 9

SAMPLE PROBLEM

● An Ophthalmologist wants to determine the mean baseline
central retinal thickness among the patients given avastin.
● To calculate for the sample size, he needs to make an estimate of the maximum allowable error of the mean
baseline central retinal thickness, i.e., ± a value of the mean estimate and a standard deviation of the baseline central retinal thickness.
● If he likes his estimate to be ±30 units of the true value, with
a reliability of 95%, and a standard deviation (SD) of 176, (this is taken from a previous study), he can calculate how many patients will be needed in study

Answer 9

133

Question 10

Study the ESTIMATION OF THE POPULATION

Answer 10

OKS

Question 11

SAMPLE PROBLEM

● A Biochemist wishes to estimate the prevalence of zinc deficiency among diabetic patients.
● He needs to get the prevalence of zinc deficiency among diabetic patients in previous studies.
● Suppose the prevalence in a related study is 10% and the desired precision (or maximum allowable error) is ±3% and a reliability of 95%, he can calculate how many participants
should be included in her study.

Answer 11

385

Question 12

● Factors to be considered in the computation of sample size
● αerror (Z1-α)
● βerror (Z1-β) = Power of the test
● Effect size (μ1–μ2)
● Type of the alternative hypothesis
● Variability as measured by the SD

Answer 12

TEST OF HYPOTHESIS

Question 13

Sample probelm

• An Ophthalmologist wants to determine if there is a difference in the percent reduction of central retinal thickness between those given less than 4 injections of avastin after one month of treatment and those given 4 or more.
• Things he need to assume, based on a previous study:

aerror = 5%
Power of the test = 95%
Effect size (N,-H,) = 10%
Type of the alternative hypothesis = 2 tailed
SD = 10 for the both groups

Answer 13

16 𝑝𝑎𝑡𝑖𝑒𝑛𝑡𝑠 𝑝𝑒𝑟 𝑔𝑟𝑜𝑢𝑝

Question 14

Sample Problem

• A hospital administrator wants to compare the incidence of neonatal infections among premature babies with one group of babies breastfed fully and the other group, bottle fed.
• Things he need to assume, based on a previous study:
• aerror = 5%
• Power of the test = 90%
• Incidence of neonatal infections among premature babies is 25%, and breastfeeding is hypothesized to reduce the incidence to
• 15% (Effect size = 10%)
• Type of the alternative hypothesis = 1 tailed
• Sample size required is 271 patients per group or 542 for two groups.

Answer 14

271 𝑝𝑎𝑡𝑖𝑒𝑛𝑡𝑠