Dear Sir

The sample from infinite population , need to know some information about population , as standard deviation , you can get it from previous study or from a pilot study , then you can use the formula :

n= [( 2* Z * σ ) / L ] ^2

where 

σ : standard deviation 

L : confidence interval for your estimation 

good luck 

Sir,

Sample size depends on a number of factors. A good book on sampling is W. G. Cochran, Sampling Techniques, John Wiley and Sons, New York, 3rd edition, 1977. Download versions exist.

Wim Voorn.

I think there is a standard method on any Research methodology book with Statistics method will solve the right way ...to decide the sample size ..

In the absence of any other information whatsoever you could always go with Cohen's d:

"For Cohen's d an effect size of 0.2 to 0.3 might be a "small" effect, around 0.5 a "medium" effect and 0.8 to infinity, a 'large' effect."

"The terms 'small,' 'medium,' and 'large' are relative, not only to each other, but to the area of behavioral science or even more particularly to the specific content and research method being employed in any given investigation....In the face of this relativity, there is a certain risk inherent in offering conventional operational definitions for these terms for use in power analysis in as diverse a field of inquiry as behavioral science. This risk is nevertheless accepted in the belief that more is to be gained than lost by supplying a common conventional frame of reference which is recommended for use only when no better basis for estimating the ES index is available."

Also check out this reference. It is a systematic review which suggests that one half a sd may be appropriate for judging clinical significance. This correlates with a cohens d of 0.5.

Interpretation of changes in health-related quality of life: the remarkable universality of half a standard deviation.

Norman GR, et al. Med Care. 2003

 Page 1 of 1 

Sample size depends upon the variability among the characteristics of the units. for example we need data on human body  temperature then we need only few samples where as data on income required a large no. of units in the samples. 

The sample size will depend on the desired confidence interval. The confidence interval for a given sample size will vary depending on the prevalence of the feature of interest. The confidence interval will be at its widest when the prevalence is 50%, so in the absence of any information it's best to set the sample size to yield the desired confidence interval around a prevalence of 50%

In simple figures:

N=30 gives a confidence interval of ±20%

N=50 gives a confidence interval of ±15%

N=100 gives a confidence interval of ±10%

N=400 gives ±5%

N=1000 gives ±3%

The Chebycheff inequality gives a bound for the confidence level of a confidence interval. But this bound depends (in the case of estimating the population mean with a sample mean) of the variance of the supposed infinite population. As an infinite population could have variance infinite, it is possible to conclude that in some cases it is not possible determine a finite sample size for a predetermined confidence level.

In the case of having population variance finite, it would be possible to find a sample size which assures a determinated confidence level.

If the variance of the infinite population is finite and known a priori, the sample size can be determined with exactness. But if the variance of the infinite population is finite and unknown, the sample size would be approximated for a predetermined confidence level, since the population variance must be approximated for the "sample quasivariance" which is an unbiased estimator of the population variance, to obtain an estimation of the searched sample size.

In practice, the population variance is not known in advance. But it is possible to estimate without bias this population variance in a pilot study, and consequently it is possible to estimate without bias the sample size for a posterior study.

Dear Sir

The sample from infinite population , need to know some information about population , as standard deviation , you can get it from previous study or from a pilot study , then you can use the formula :

n= [( 2* Z * σ ) / L ] ^2

where 

σ : standard deviation 

L : confidence interval for your estimation 

good luck 

the other solution, but needs scientific background explanation 

http://www.williamgodden.com/samplesizeformula.pdf

If the population size in infinity then 9604 Within (+/ - 1%) and 384 (+/ - 5%) by Teddlie et.al. (2009).