Scribd
Upload
13 views

Sample Size Calculation

Uploaded by

kumarianuradha125.ak
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
13 views

Sample Size Calculation

Uploaded by

kumarianuradha125.ak
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 14

Sample Size Calculation

Sample Size Determination


Sample size refers to the number of individuals included in a
research study as a representative of the total population.
The sample size is determined from the target study population.
Sample size determination is the most crucial methodological part
of a research study.
It is essential for an investigator to estimate optimum sample size to
produce reliable results.
The bigger the sample size, the better the results of the research.
The determination of appropriate sample size is the most
important factor for the estimation of population
parameters, i.e. for the estimation of population mean,
population proportion or population standard deviation.

The important methods of determination of sample size are


on the basis of following two conditions.
i. For quantitative
ii. For qualitative
How to Calculate Sample Size
• Determine the population size ( if known).

• Determine the confidence interval. (either 95% or 99%)

• Determine the confidence level. ( alpha value 0.05 or 0.01).

• Determine the standard deviation ( a standard deviation of 0.5 is a safe


choice where the figure is unknown).

• Convert the confidence level into a Z -score.


Why do you need to calculate sample size

• Proper precision
• Proper level of confidence
• Adequate power
• Address affect size

• Very large sample size: Disadvantages


• Waste of limited resource (manpower, time, money, instruments)
• Recruiting more subject than required is unethical
• In RCTs, large number of subject will get placebo or inferior treatment (unethical)
Sample size estimation for cross-sectional descriptive study
For Quantitative/Numeric data
For infinite population
n= , Where n= sample size, d= margin of error or allowable error (5%)
Z= level of significance (1.96 at 5% level of significance)
= standard deviation (SD)
For finite population
N=

The mean systolic blood pressure of certain rural community was found
to be 122 mm of Hg with standard deviation of 12mm of Hg. Calculate the
minimum sample size to verify the result if allowable error is at 5% risk. If
there were 400 household , what should be the sample size at 95% CI?
 The mean systolic blood pressure of patients of a hospital
was found 120mm of Hg with a standard deviation of 10
mm of Hg. Calculate minimum sample size, if allowable
error is 2% of mean at 5% risk.
Solution, Population mean (ն) = 120 mm of Hg
Population S.D. (ճ) = 10 mm of Hg
Allowable error (E) = 2% of mean = 2/100*120= 2.4
Alpha =5% = 0.05
Z = 1.96
We have, the required sample size (n) is,
n = = = 66.69 ֘ 67
FOR QUALITATIVE DATA
In such data, proportion(P) of the attribute is calculated.
For finding the sample size, the assumption usually made is that the allowable error
does not exceed 10% or 20% of the positive character(P).
The sample size is determined with a desired margin of error at 5% risk that the true
estimate will not exceed allowable error by 10% or 20% of P.
For populations that are large,
Cochran developed the equation to yield a representative sample for proportions.
n=
Where n = sample size,
Z= level of significance (1.96 at 5% level of significance)
d= margin of error
p is the estimated proportion of an attribute or prevalence present .
q = 1-p
 In a study of rice eaters in Bharatpur 70% of the population is supposed
to be rice eaters with 95% confidence level. What is the sample required
size to verify the supposition so that the sampling error is not more that
5% below and above the true proportion of rice eaters. Ans 316
FINITE POPULATION CORRECTION FOR PROPORTIONS

If the population is small then sample size can be reduced


slightly. This is because a given sample size provides
proportionately more information for a small population than
for large population. The sample size() can be adjusted by using

n=
Where n is the sample size and N is the population size.
SIMPLIFIED FORMULA FOR PROPORTION

Yamane (1967) provides a simplified formula to calculate sample size when the
population size(N) is given only. This formula is used to calculate the sample
sizes for 95% confidence level and p=0.5 is assumed.
n=

Where n= sample size, N= known population of the study & e= allowable


error.
• A researcher wishes to find the minimum sample size to study the prevalence of
smokers in the adolescent group in a certain college. The total number of
students of that college was 2000. How much sample size requires to him to
conduct the research? Ans 333
n=
Where e= allowable error (5%)
Z= level of significance (1.96 at 5% level of significance)
N = Total number of patient related to that disease in any two disease.
P= probability (0.5)
Factors affecting sample size
1. Design effect
2. Resources available
3. Nature of study
4. Sampling methods used
5. Homogeneity
6. Effect size
7. Degree of confidence.
8. Measurement factors
 Define sampling and Non-sampling errors.
Write how to minimize sampling error.

You might also like