Sample size-2
Choosing the size of your sample is related to your expected signal-to-noise ratio and your desired confidence.
Statisticians speak about statistical power, for details see [Trochim2010] http://www.socialresearchmethods.net/kb/power.php Links to an external site.
See also: [Kelley2008] [Maxwell2008 [Kelley2003a] [Kelley2003b] [Kelley2008a]
[Trochim2010] William M.K. Trochim, ‘Statistical Power’, 20-Oct-2010. [Online]. Available: http://www.socialresearchmethods.net/kb/power.php Links to an external site. . [Accessed: 03-Aug-2015]
[Kelley2008] Ken Kelley and Scott E. Maxwell, ‘Sample Size Planning with Applications to Multiple Regression: Power and Accuracy for Omnibus and Targeted Effects’, in The SAGE handbook of social research methods, Pertti Alasuutari, Ed. Newbury Park, CA, USA: SAGE Publications Ltd, 2008, pp. 166–192 [Online]. Available: http://nd.edu/~kkelley/publications/chapters/Kelley_Maxwell_Chapter_SSMR_2008.pdf Links to an external site.
[Maxwell2008] Scott E. Maxwell, Ken Kelley, and Joseph R. Rausch, ‘Sample size planning for statistical power and accuracy in parameter estimation. Annual Review of Psychology’, Annual Review of Psychology, vol. 59, pp. 537–563, 2008. DOI: 10.1146/annurev.psych.59.103006.093735 http://nd.edu/~kkelley/publications/articles/Maxwell_Kelley_Rausch_2008.pdf Links to an external site.
[Kelley2003a] Ken Kelley and Scott E. Maxwell, ‘Sample Size for Multiple Regression: Obtaining Regression Coefficients That Are Accurate, Not Simply Significant.’, Psychological Methods, vol. 8, no. 3, pp. 305–321, 2003. DOI: 10.1037/1082-989X.8.3.305 http://nd.edu/~kkelley/publications/articles/Kelley_Maxwell_2003.pdf Links to an external site.
[Kelley2003b] Ken Kelley, Scott E. Maxwell, and Joseph R. Rausch, ‘Obtaining Power or Obtaining Precision: Delineating Methods of Sample-Size Planning’, Evaluation & the Health Professions, vol. 26, no. 3, pp. 258–287, Sep. 2003. DOI: 10.1177/0163278703255242 http://nd.edu/~kkelley/publications/articles/Kelley_Maxwell_Rausch_2003.pdf Links to an external site.
[Kelley2008a] Ken Kelley, Keke Lai, and Po-Ju Wu, ‘Ch. 34: Using R for data analysis: A best practice for research’, in Best practices in quantitative methods, Jason W. Osborne, Ed. Newbury Park, CA, USA: SAGE Publications, Inc., 2008, pp. 535–572. http://nd.edu/~kkelley/publications/chapters/Kelley_Lai_Wu_Using_R_2008.pdf Links to an external site.
Transcript
So back to the problem about size. Now, statisticians talk about statistical power, and for details, you can read this reference at http://www.socialresearchmethods.net/kb/power.php. But basically, you can think about is: What are we trying to do? We want to choose a sample size so that we can get our signal large enough above the expected noise that we can see our signal - with some given level of confidence.